Continental and global hydrological models are the primary means to simulate surface/sub-surface water storage, water flux, and surface water inundation variables, which are required for hazard mitigation and policy support plans. However, establishing these large-scale models is challenging since the complicated physical processes that govern large-scale hydrology cannot be fully resolved by the simplified equations in these schemes. Besides, it is well known that the model parameters are insufficient to account for intensification of the water cycle caused by the climate change and anthropogenic modifications. Another issue is that most hydrological and hydraulic models are at best only calibrated against river discharge or similar data, but these calibrated parameters may have limited influence on the estimation of water storage and water volume changes in large-scale basins. In this study, we demonstrate the extent to which Terrestrial Water Storage (TWS; a vertical summation of surface and sub-surface water storage) data from the Gravity Recovery And Climate Experiment (GRACE) and its follow-on mission (GRACE-FO), as well as remotely sensed soil moisture data can improve the estimations of river discharge and water extent as well as water storage during episodic droughts and floods. For this, we present the structure of our in-house ensemble Kalman filter based calibration and data assimilation (C/DA) as well Bayesian model-data merging frameworks to integrate freely available satellite data into the in-house modified W3RA water balance model forced by ERA5 data. The results are demonstrated through simulations of water storage, river discharge, drought characteristics and floods in West Africa and Europe.