@article{10.1029/2018SW002110,
author = {Aseev, N. A. and Shprits, Y. Y.},
title = {Reanalysis of Ring Current Electron Phase Space Densities Using Van Allen Probe Observations, Convection Model, and Log-Normal Kalman Filter},
journal = {Space Weather},
volume = {17},
number = {4},
pages = {619-638},
keywords = {ring current, data assimilation, inner magnetosphere, reanalysis, Kalman filter},
doi = {https://doi.org/10.1029/2018SW002110},
url = {https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2018SW002110},
eprint = {https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2018SW002110},
abstract = {Abstract Models of ring current electron dynamics unavoidably contain uncertainties in boundary conditions, electric and magnetic fields, electron scattering rates, and plasmapause location. Model errors can accumulate with time and result in significant deviations of model predictions from observations. Data assimilation offers useful tools which can combine physics-based models and measurements to improve model predictions. In this study, we systematically analyze performance of the Kalman filter applied to a log-transformed convection model of ring current electrons and Van Allen Probe data. We consider long-term dynamics of μ = 2.3 MeV/G and K = 0.3 G1/2RE electrons from 1 February 2013 to 16 June 2013. By using synthetic data, we show that the Kalman filter is capable of correcting errors in model predictions associated with uncertainties in electron lifetimes, boundary conditions, and convection electric fields. We demonstrate that reanalysis retains features which cannot be fully reproduced by the convection model such as storm-time earthward propagation of the electrons down to 2.5 RE. The Kalman filter can adjust model predictions to satellite measurements even in regions where data are not available. We show that the Kalman filter can adjust model predictions in accordance with observations for μ = 0.1, 2.3, and 9.9 MeV/G and constant K = 0.3 G1/2RE electrons. The results of this study demonstrate that data assimilation can improve performance of ring current models, better quantify model uncertainties, and help deeper understand the physics of the ring current particles.},
year = {2019}
}