EnKF Ionosphere and Thermosphere Data Assimilation Algorithm Through a Sparse Matrix Method

• Published in: Journal of geophysical research. Space physics 2019-08, Vol.124 (8), p.7356-7365
• Main Authors: He, Jianhui, Yue, Xinan, Wang, Wenbin, Wan, Weixing
• Format: Article
• Language: English
• Subjects: Algorithms; Atmospheric circulation; Atmospheric research; Computer simulation; Covariance; Data assimilation; Data collection; Electrodynamics; Electron density; General circulation models; Ionosphere; Kalman filters; Mathematical models; Optimization; Performance assessment; Reliability analysis; Sparse matrices; Sparsity; State variable; Studies; Supercomputers; Thermosphere; Work stations; Workstations
• ISSN: 2169-9380; 2169-9402
• DOI: 10.1029/2019JA026554

In this work, we constructed an ensemble Kalman filter (EnKF) ionosphere and thermosphere data assimilation system using the National Center for Atmospheric Research Thermosphere Ionosphere Electrodynamics General Circulation Model (NCAR‐TIEGCM) as the background model. We use a sparse matrix method to avoid significant matrix related calculation and storage. A series of observing system simulation experiments have been conducted to assess the performance of the system. The results show that the system optimizes ionosphere drivers efficiently by assimilating electron densities through their covariance. The short‐term forecast capability is enhanced significantly, and the effect of initial condition correction lasts for longer than 24 hr. To our knowledge, this is the first study to demonstrate that the EnKF‐based global ionosphere and thermosphere data assimilation can be conducted without using a supercomputer. This workstation‐based EnKF ionosphere and thermosphere data assimilation system benefits both scientific studies and near‐real‐time operation. Plain Language Summary In the ionosphere and thermosphere, the neutrals remember the past much longer than the ionized part does. In the recent decades, ensemble‐based data assimilation has been proven to be an efficient method to enhance the ionosphere forecast capability by assimilating electron densities through optimizing neutral state variables via their covariance. Due to the globality and large ensemble running, this kind of studies always relies on supercomputers. In this study, we used a sparse matrix method to do the matrix‐related calculation and storage in the EnKF and conducted EnKF ionosphere and thermosphere data assimilation on a workstation. We have done a series of observing system simulation experiment studies to evaluate the validity and reliability of the method. The results show that the EnKF can optimize the ionosphere drivers efficiently by assimilating electron density through their covariance. The short‐term forecast capability is enhanced significantly and extended to last longer than 24 hr. This is the first study to demonstrate that the EnKF‐based global ionosphere and thermosphere data assimilation system can be conducted without using supercomputers. It can benefit both scientific studies using the EnKF ionosphere and thermosphere‐related data assimilation system and the near‐real‐time operational purpose. Key Points A new ionosphere and thermosphere data assimilation system i