CoAuthors

Anny Cazenave (LEGOS-CNES, France); Hindumathi Palanisamy (LEGOS-CNES, France); Benoit Meyssignac (LEGOS-CNES, France)

The satellite altimetry-based global mean sea level (GMSL) evolution is compared to the sum of the steric and mass contributions over the 2003-2010/12 time span. Different solutions are considered for the GMSL: AVISO, NOAA, Colorado University, GSFC, CSIRO and ESA CCI. The mass component is estimated from GRACE (data from CSR, GFZ and JPL processed by D. Chambers). For the steric component, we used different data sets, including 4 different Argo products. We quantify the uncertainty of each component (i.e., GMSL, steric and mass components) from the dispersion between the different products around the mean. Large uncertainties are noticed at interannual time scale while trends tend to better agree. In principle, non closure of the sea level budget is able to provide some constraint on the deep ocean steric contribution (below 1500 m). This is discussed in the context of the various data sets used in this analysis. We also examine different ocean reanalyses (e.g., ORAS4, SODA, ...). We compare the ocean reanalyses with Argo for the upper 1500 m and also estimate the deep ocean steric contribution as the reanalyses provide information over the whole ocean depth range. Significant differences are observed between the reanalysis-based steric grids in some regions, depth layers, and time spans. We discuss all these results and their ability to provide constraints on deep ocean warming over the last decade.