Application of a new data operator-splitting data assimilation technique to the 3-D VERB diffusion code and CRRES measurements

• Published in: Geophysical research letters Vol. 40; no. 19; pp. 4998 - 5002
• Main Authors: Shprits, Yuri, Kellerman, Adam, Kondrashov, Dmitri, Subbotin, Dmitriy
• Format: Journal Article
• Language: English
• Published: Washington Blackwell Publishing Ltd 16-10-2013; John Wiley & Sons, Inc
• Subjects: Data assimilation; Data collection; Geophysics; Kalman filters; modeling; Radiation; radiation belts
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1002/grl.50969

In this study we present 3‐D data assimilation using CRRES data and 3‐D Versatile Electron Radiation Belt Model (VERB) using a newly developed operator‐splitting method. Simulations with synthetic data show that the operator‐splitting Kalman filtering technique proposed in this study can successfully reconstruct the underlying dynamic evolution of the radiation belts. The method is further verified by the comparison with the conventional Kalman filter. We applied the new approach to 3‐D data assimilation of real data to globally reconstruct the dynamics of the radiation belts using pitch angle, energy, and L shell dependent CRRES observations. An L shell time cross section of the global data assimilation results for nearly equatorially mirroring particles and high and low values of the first adiabatic invariants clearly show the difference between the radial profiles of phase space density. At μ = 700 MeV/G cross section of the global reanalysis shows a clear peak in the phase space density, while at lower energy of 70 MeV/G the profiles are monotonic. Since the radial profiles are obtained from one global reanalysis, the differences in the profiles reflect the differences in the underlying physical processes responsible for the dynamic evolution of the radiation belt energetic and relativistic electrons.
Key Points
Operator‐splitting 3‐D reanalysis can accurately reconstruct PSD
Profiles of PSD at lower energies are monotonic
Three‐dimensional reconstruction shows clear peaks in PSD MeV