Reduction of ocean tide aliasing errors in GRACE, GRACE-FO, and future mission gravity field recovery

The GRACE and GRACE-FO missions have been fundamental in establishing a near-continuous time series of global mass transport since 2002. However, monthly gravity field recovery using these mission data includes errors limiting the spatial and temporal resolution of the estimated gravity field solutions. Besides the noise of the accelerometer instruments, the primary error contributions arise from temporal aliasing errors due to undersampling of signals to be recovered (e.g., hydrology), uncertainties in the de-aliasing models (e.g., non-tidal atmosphere and ocean), and imperfect ocean tide models. Especially the latter will also remain one of the most limiting factors in determining high-resolution temporal gravity fields from Next-Generation Gravity Missions (NGGM).

In this context, recent further developments have been made within the research unit, NEROGRAV (New Refined Observations of Climate Change from Spaceborne Gravity Missions), funded by the German Research Foundation (DFG). One of the NEROGRAV projects deals with stochastic modeling of ocean tide background models utilizing covariance information for eight major tidal constituents. The repeatable pattern of the tidal signal enables the extraction of uncertainty information by an ensemble of different ocean tide models. This information can be introduced into the gravity field recovery process as a covariance matrix while expanding the parameter space by additional tidal parameters to be estimated.

This presentation provides an overview of the recovered monthly gravity fields from GRACE/GRACE-FO when applying covariance information of ocean tide errors. In addition, realistic simulations have been performed to assess the pure effect of the covariance information on ocean tide errors and the potential of co-estimating ocean tides over a longer period while considering tidal covariance information is addressed. It is shown that the application of ocean tide covariance information contributes to the reduction of temporal aliasing caused by the mismodeling of ocean tide background models.