Determination of homogenous GRACE FOLLOW-ON monthly displacements based on available solutions

Presently, Gravity Recovery and Climate Experiment (GRACE) mission data is widely used in various fields of science. The longest satellite gravimetric mission regularly explored changes of the gravity field from April 2002 to October 2017. Nowadays, its follow-on mission (GRACE-FO) observes gravity changes from May 2018, providing new greater research opportunities. In the following research, we present a completely new vertical deformations changes model of the first 14 months GRACE-FO observations. The study’s aim is to reduce the signal noise left after Gauss spatial smoothing. In this study, we use monthly gravity field in spherical harmonics form up to degree and order 96, provided by three different centers, i.e. the NASA’s Jet Propulsion Laboratory (JPL), the German Research Center for Geosciences (GFZ) and the Center for Space Research (CSR). In following study, we use all sets of data (JPL, GFZ and CSR) to test three various algorithms: (1) coefficient-wise and (2) field-wise non-iterative weighting methods and further, (3) estimation by iterative variance component method.  Finally, we obtain joined spherical harmonics changes by a weighted average scheme, which are converted to Earth crust vertical deformations. The used weighting methods reduce root mean square scatter of monthly deformation fields by nearly 5% for continental areas (excluding Amazon basin and Hudson Bay region) and by 10-15% changes for the ocean areas. However, with respect to each new-created model, differences in monthly deformation changes are in the range from ±3 mm to ±6 mm depending on the data center (JPL, GFZ or CSR). Furthermore, the analysis of signal information contained in each degree allows us to assess the quality of the created models. The highest signal variations in our models occur up to the degree and order 25. Additionally, the high differences in the signal are obtained for sectoral harmonics up to the maximum degree. Analysis showed that the applied field-wise weights much more effectively remove remaining noise after spatial averaging than per-order/degree weighting. Whereas, the obtained results indicate that observations provided by GFZ center have the smallest weights for each algorithm.