Sequential calibration and data assimilation for predicting atmospheric variability

Estimating global and multi-level variations of the atmospheric variables and being able to predict them are very important for studying coupling processes within the atmosphere, and for various geodetic and space weather applications. These variables include the thermosphere neutral density, the ionospheric electron density, and the tropospheric water vapour, which are relevant to applications such as orbit determination, satellite navigation, and weather/climate monitoring. Available models have difficulties in realistic prediction of these variables due to the simplicity of their structure or sampling limitations. In this study, we present an ensemble-based simultaneous Calibration and Data Assimilation (C/DA) algorithm to integrate freely available satellite geodetic data (e.g., CHAMP, GRACE(-FO), Swarm, and GNSS) into empirical models with the focus on improving the predictability of atmospheric variables. The improved model, called `C/DA-model' will be assessed in relevant geodetic and space weather applications. For demonstration, the CDA-NRLMSISE-00 is examined during seven periods with relatively high geomagnetic activity and CDA-IRI-ZWD during extensive rainy events.