Empirical Data Assimilation for Merging Total Electron Content Data with Empirical and Physical Models

An accurate estimation of ionospheric variables such as the total electron content (TEC) is important for many space weather, communication, and satellite geodetic applications. Empirical physics-based models are often used to determine TEC in these However, it known that cannot reproduce all variability due various reasons their simplified model structure, coarse sampling inputs, dependencies calibration period. Bayesian-based data assimilation (DA) techniques improving model’s performance, but computational cost considerably large. In this study, first, we review available DA upper atmosphere assimilation. Then, will present an empirical decomposition-based (DDA), based on principal component analysis ensemble Kalman filter. DDA reduces complexity previous implementations. Its performance demonstrated by updating orthogonal functions NeQuick TIEGCM using rapid global ionosphere map (GIM) products observation. The new models, respectively, called ‘DDA-NeQuick’ ‘DDA-TIEGCM,’ then predict values next day. Comparisons forecasts with final GIM (that after 11 days) represent average $$42.46\%$$ $$31.89\%$$ root mean squared error (RMSE) reduction during our test period, September 2017.