Thermospheric mass density specification using an ensemble Kalman filter

Tomoko Matsuo, I. Te Lee, Jeffrey L. Anderson

Research output: Contribution to journalArticlepeer-review

47 Scopus citations


This paper presents an application of ensemble Kalman filtering (EnKF) to a general circulation model of the thermosphere and ionosphere. It is designed to incorporate the feedback between plasma and neutral variables in both the analysis and forecast steps of filtering so that thermospheric parameters can be inferred from ionospheric observations and vice versa. We make a case that the global neutral density specification can greatly benefit from this approach based on a number of filtering experiments conducted under the assumption of no model bias. Specific observations considered are (i) neutral mass densities obtained from the accelerometer experiment on board the CHAMP satellite and (ii) electron density profiles obtained from the COSMIC/FORMOSAT-3 mission. Assimilation of the neutral mass density obtained from the CHAMP mission into the TIEGCM is shown to improve the neutral density specification in the vicinity of satellite orbits, but is short of making a global impact unless accompanied by the estimation of the primary driver of the density variability such as solar EUV flux. On the other hand, assimilation of the COSMIC/FORMOSAT-3 electron density profiles into the Thermosphere-Ionosphere-Electrodynamics General Circulation Model (TIEGCM) is far more effective than the CHAMP neutral density in terms of improving the global neutral mass density specification. This suggests a synthesis of thermospheric and ionospheric observations into the general circulation model, brought about with the help of the latest EnKF techniques, can effectively increase the geophysical information content of observations. Key Points This technique can increase geophysical information content of observations Possible to infer thermospheric parameters from electron density observations The EnKF system was constructed with NCAR community softwares

Original languageEnglish
Pages (from-to)1339-1350
Number of pages12
JournalJournal of Geophysical Research: Space Physics
Issue number3
StatePublished - Mar 2013


  • data assimilation
  • ionosphere
  • thermosphere


Dive into the research topics of 'Thermospheric mass density specification using an ensemble Kalman filter'. Together they form a unique fingerprint.

Cite this