Estimation of the topside ionosphere and plasmasphere by an Ensemble Kalman Filter and the simultaneous multiplicative column-normalized method SMART+

Trans-ionospheric radio signals, transmitted from satellites used in telecommunication, navigation or remote sensing, are affected by the electron density along their ray path through the plasmasphere and ionosphere. Especially in the nighttime, up to 50% of the signal delays of L-band signals used in the Global Navigation Satellite System may originate from altitudes above ionospheric F2 layer, which consists of the topside ionosphere going over into the plasmasphere. In general, the majority of data, available for the reconstruction of the ionosphere and plasmasphere, are Slant Total Electron Content (STEC) measurements, i.e. the integral of the electron density along the ray path between the transmitting satellite and the receiver. However, the ground based STEC values are usually dominated by the high electron density values within and below the F2 layer peak, and thus the exact information about the topside ionosphere and especially the plasmasphere is almost impossible to extract. Thus, in the presented work we concentrate on the estimation of the topside part of the ionosphere and the plasmasphere by means of space based STEC measurements only. The STEC measurements of eleven Low Earth Orbit (LEO) satellites are used for the reconstructions by an Ensemble Kalman Filter (EnKF) and the Simultaneous Multiplicative Column-Normalized Method (SMART+), where the NeQuick model serves as background. We investigate the capability of the estimations to reconstruct the assimilated STEC, independent STEC measurements, Swarm Langmuir Probes (LP) electron density measurements and COSMIC Ionospheric Radio Occultation (IRO) profiles for two periods of the year 2015 covering moderate to perturbed ionospheric conditions. We observe that in median the NeQuick model underestimates the IRO electron densities in the moderate period and for high latitudes in the perturbed period as well as the STEC measurements for the three reference LEO satellites, GRACE and Swarm A, during both periods. The comparison with the assimilated STEC shows that during both periods both methods reduce the STEC residuals by up to 86% compared to the background model NeQuick. The residuals of the independent STEC of the two methods are up to 64% smaller compared to the NeQuick model residuals. For the NeQuick model the median of the absolute relative IRO residuals is about 33% in the quiet period and about 25% in the perturbed period. SMART+ reduces these residuals whereas EnKF does not provide an improvement.