Future satellite gravity field missions – Impact of a direct time-variable parametrization

The limited achievable temporal resolution poses one of the main limitations of current satellite gravity field missions such as GRACE and GRACE-FO and is, eventually, responsible for temporal aliasing. Increasing the temporal resolution is thus one of the most important tasks for future satellite gravity field missions. This, however, can only be achieved by means of larger satellite constellations since the temporal resolution can basically be defined as the time needed to achieve a global observation coverage. This needed (retrieval) time scales linearly with the number of satellites: e.g., a two-pair mission such as the upcoming joint ESA/NASA MAGIC mission can easily achieve global coverage with sufficient spatial resolution within less than a week. For a hypothetical 6-pair mission, even an independent daily retrieval is feasible. Future missions will hence allow to sample the gravity field in much shorter intervals. However, up until now, the parametrization of the gravity field within these intervals usually only accounts for a static behavior, resembling a step function in the time domain. Obviously, such a behavior is unnatural and does not follow the actual progression of the gravity field. So, even if sufficient temporal resolution would be available, temporal aliasing will not be fully mitigated due to this mis-parametrization. In this contribution we will thus investigate the impact of a direct time-variable parametrization through continuous spline-functions. On the example of a fictive 6-pair mission, we show how such a spline parametrization with daily support points can be applied: firstly, we prove that the spline parametrization allows numerically stable and correct solution based on a closed loop scenario without residual (sub-daily) temporal aliasing. Secondly, based on a more realistic scenario with sub-daily temporal gravity signal, we also highlight the (theoretical) limitations of a 6-pair mission due to the still very dominant temporal aliasing (from sub-daily signal sources). This work is supported by the ESA QSG4EMT study in collaboration with Politecnico di Milano, Delft University of Technology, HafenCity University Hamburg, University of Bonn and University of Trieste.