Item

Abstract

We assess GRACE Follow-On (GFO) level-2 data products (monthly sets of spherical harmonic coefficients) produced by different data processing teams. To that end, we apply the technique proposed by Ditmar (2022) to the time-series of mass anomalies computed at the nodes of a global grid. Noise levels of the time-series are estimated simultaneously, node-by-node. The procedure is based on a production of a common mass anomaly time-series in the presence of a dedicated regularization functional. Noise levels in the individual time-series are estimated iteratively by means of Variance Component Estimation.We report average noise levels calculated both globally and over specific geographical areas, including those with a strong signal (e.g., ice sheets). Furthermore, we compare the commonly used solutions based on K-Band ranging (KBR) with experimental solutions based on Laser Ranging Interferometry (LRI) data. In addition, we compare the performance of all those solutions with the latest variant of solutions based on GRACE data.Our results will allow the community of GRACE/GFO data users to make a motivated choice of the data product(s) to be used as input in their studies. Additionally, the produced estimates can help the developers of level-2 GRACE/GFO data products to identify weak points of their solutions that manifest themselves as areas of a relatively high noise level. Those may reveal, e.g., deficiencies of the exploited ocean tide models at the identified areas. This will pave the way for a further improvement of data products based on satellite gravimetry data from GRACE and GFO missions.