Inferring Long-Term Geocenter Motion From Low-Degree Gravity Field

Accurate determination of geocenter motion is essential not only for establishing a stable terrestrial reference frame, but also for deriving a complete picture of large-scale global mass redistribution in the Earth system. For geophysical applications, reliable geocenter motions can be inferred from time-variable gravity fields provided by the Gravity Recovery and Climate Experiment (GRACE) since 2002, but it is more challenging for the pre-GRACE era where only low-degree gravity fields are available from the Satellite Laser Ranging (SLR). In addition, geocenter motion estimates derived from SLR using the direct method lack the trend in a linear reference frame and are therefore generally not suitable for studying mass change rates. In this study, we derive the geocenter motion from low-degree gravity fields up to degree and order 5 after properly addressing signal leakage. Using the leakage-corrected land mass patterns combined with corresponding ocean mass fingerprints, we generate geocenter motion estimates and compare them with those derived from GRACE, geophysical models, and the SLR direct method between 2002 and 2020. The trends in our estimates are consistent with GRACE and models, with differences below 0.1∼0.2 mm/yr depending on the quality of the gravity field models, while the SLR direct estimates yield opposite trends, leading to significantly underestimated global ocean mass change rates. Our study provides the first promising solution to derive long-term geocenter motion rates from low-degree gravity fields, which can be used to track large-scale mass change back to the 1990s.