Ensemble–Variational Integrated Localized Data Assimilation

TLDR

Hybrid variational–ensemble data assimilation is the current state of the art for initializing numerical weather prediction models, but it requires a separate ensemble DA to estimate uncertainty, which can be suboptimal technically and scientifically. EVIL aims to generalize ensemble Kalman filter methods within a variational framework, updating ensemble analyses with deterministic variational information while focusing on affordability and efficiency for operational use. EVIL updates ensemble analyses with deterministic variational information, employing a localized integrated approach designed for affordability and operational efficiency.

Abstract

Abstract Hybrid variational–ensemble data assimilation (hybrid DA) is widely used in research and operational systems, and it is considered the current state of the art for the initialization of numerical weather prediction models. However, hybrid DA requires a separate ensemble DA to estimate the uncertainty in the deterministic variational DA, which can be suboptimal both technically and scientifically. A new framework called the ensemble–variational integrated localized (EVIL) data assimilation addresses this inconvenience by updating the ensemble analyses using information from the variational deterministic system. The goal of EVIL is to encompass and generalize existing ensemble Kalman filter methods in a variational framework. Particular attention is devoted to the affordability and efficiency of the algorithm in preparation for operational applications.