Using stochastic information on geophysical processes for daily GRACE Kalman filter solutions

The Research Unit (RU) New Refined Observations of Climate Change from Spaceborne Gravity Missions (NEROGRAV) aims at improving the understanding and providing new information about sensor data, background models and the processing strategies for the GRACE and GRACE-FO data. The achieved improvements in the standard GRACE data processing create a possibility for improvements in the computation of gravity fields with a high temporal resolution. In the current contribution we focus on the studying the potential of daily gravity filed solutions obtained by Kalman filtering. We use in particular the NEROGRAV results on stochastic modeling of the observations and background models used in the GRACE data processing, as well as the provided stochastic information on the geophysical processes governing the variations of the gravity field on sub-monthly time scales.

 

While the standard GRACE gravity field solutions are obtained using one month of observational data, there are on-going attempts to obtain high-quality gravity fields with a higher temporal resolution. Unlike the monthly solutions, which are computed fully independent of each other, the computation of the daily gravity fields requires taking into account the temporal correlations between the subsequent solutions. The assessment of this timely correlation is based on the knowledge of the stochastic properties of the underlying geophysical processes. In this contribution we study how the daily Kalman solutions are influenced by the stochastic information on the two main geophysical processes causing high-frequency variations in the gravity field: non-tidal atmospheric and oceanic variations and hydrology. We present results for daily gravity field solutions for several test months in the spectral and spatial domain.