Atmospheric Contributions to Global Ocean Tides for Satellite Gravimetry

To mitigate temporal aliasing effects in monthly mean global gravity fields from the GRACE and GRACE-FO satellite tandem missions, both tidal and non-tidal background models describing high-frequency mass variability in atmosphere and oceans are needed. To quantify tides in the atmosphere, we exploit the higher spatial (31 km) and temporal (1 hr) resolution provided by the latest atmospheric ECMWF reanalysis, ERA5. The oceanic response to atmospheric tides is subsequently modeled with the general ocean circulation model MPIOM (in a recently revised TP10L40 configuration that includes the feedback of self-attraction and loading to the momentum equations and has an improved bathymetry around Antarctica) as well as the shallow water model TiME (employing a much higher spatial resolution and more elaborate tidal dissipation than MPIOM). Both ocean models consider jointly the effects of atmospheric pressure variations and surface wind stress. We present the characteristics of 16 waves beating at frequencies in the 1-6 cpd band and find that TiME typically outperforms the corresponding results from MPIOM and also FES2014b as measured from comparisons with tide gauge data. Moreover, we note improvements in GRACE-FO laser ranging interferometer range-acceleration pre-fit residuals when employing the ocean tide solutions from TiME, in particular, for the S-1 spectral line with most notable improvements around Australia, India, and the northern part of South America.

Automated summary

To mitigate temporal aliasing effects in monthly mean global gravity fields, both tidal and non-tidal background models are needed. In this study, atmospheric tides are quantified using the latest ECMWF reanalysis ERA5, while oceanic response to atmospheric tides is modeled using the general ocean circulation model MPIOM and the shallow water model TiME. The results show that TiME outperforms MPIOM and FES2014b in modeling atmospheric tides and has notable improvements in GRACE-FO laser ranging interferometer range-acceleration pre-fit residuals.