Quantifying barystatic sea-level change from satellite altimetry, GRACE and Argo observations over 2005-2016

Time-varying spherical harmonic coefficients determined from the Gravity Recovery and Climate Experiment (GRACE) data provide a valuable source of information about the water mass exchange that is the main contributor to the Earth's gravity field changes within a period of less than several hundred years. Moreover, by measuring seawater temperature and salinity at different layers of ocean depth, Argo floats help to measure the steric component of global mean sea level (GMSL). In this study, we quantify the rate of barystatic sea-level change using both GRACE RL05 and RL06 monthly gravity field models and compare the results with estimates achieved from a GMSL budget closure approach. Our satellite altimetry-based results show a trend of 3.90 +/- 0.14 mm yr(-1) for the GMSL rise. About 35% or 1.29 +/- 0.07 mm yr(-1) of this rate is caused by the thermosteric contribution, while the remainder is mainly due to the barystatic contribution. Our results confirm that the choice of decorrelation filters does not play a significant role in quantifying the global barystatic sea-level change, and spatial filtering may not be needed. GRACE RL05 and RL06 solutions result in the barystatic sea-level change trends of 2.19 +/- 0.13 mm yr(-1) and 2.25 +/- 0.16 mm yr(-1), respectively. Accordingly, the residual trend, defined as the difference between the altimetry-derived GMSL and sum of the steric and barystatic components, amounts to 0.51 +/- 0.51 and 0.45 +/- 0.44 mm yr(-1) for RL05 and RL06-based barystatic sea-level changes, respectively, over January 2005 to December 2016. The exclusion of the halosteric component results in a lower residual trend of about 0.36 +/- 0.46 mm yr(-1) over the same period, which suggests a sea-level budget closed within the uncertainty. This could be a confirmation on a high level of salinity bias particularly after about 2015. Moreover, considering the assumption that the GRACE-based barystatic component includes all mass change signals, the rather large residual trend could be attributed to an additional contribution from the deep ocean, where salinity and temperature cannot be monitored by the current observing systems. The errors from various sources, including the model-based Glacial Isostatic Adjustment signal, independent estimation of geocenter motion that are not quantified in the GRACE solutions, as well as the uncertainty of the second degree of zonal spherical harmonic coefficients, are other possible contributors to the residual trend. (C) 2020 COSPAR. Published by Elsevier Ltd. All rights reserved.