Evaluation of the CMCC eddy-permitting global ocean physical reanalysis system (C-GLORS, 1982-2012) and its assimilation components

Ocean reanalyses are data assimilative simulations, aimed at estimating the four-dimensional state of the ocean over long periods, in a way as consistent over time as possible. They are designed for a wide range of climate applications, such as climate monitoring and low-frequency variability studies, along with several downstream applications (e.g. biogeochemical and fishery modelling, initial conditions for long-range coupled predictions, regional model nesting). An upgraded version of the Euro-Mediterranean Center for Climate Change (CMCC) eddy-permitting global ocean reanalysis, named CMCC Global Ocean Reanalysis System (C-GLORS) version 4, was recently released. The reanalysis covers the meteorological satellite era (1982-2012). This article details the configuration of the reanalysis system and provides an extensive validation, focusing on the evaluation of main indexes related to climate monitoring. Cumulative denial experiments are also conducted, in order to understand the relative impact of assimilation components included in C-GLORS (i.e. altimetric data, variational assimilation, bias correction and surface nudging). Results indicate that C-GLORS proves reliable in simulating long-term means, heat and freshwater trends, sea-level variability, mean surface circulation and transports, eddy variability and meridional overturning circulation and its associated heat transport, except for a few specific issues (overestimation of volume transports in the Southern Ocean and slight underestimation of the Atlantic ocean meridional overturning circulation and associated heat transport, the latter mostly linked to underestimation of western boundary northward transports). The results also demonstrate the complementarity of the assimilation components, all improving verification skill scores, for example the importance of the variational assimilation for the simulation of the reanalysis small-scale variability, the importance of the bias-correction scheme for correcting subsurface salinity errors or the role of surface nudging in driving the North Atlantic ocean circulation. The analyses presented here offer ideas for improving C-GLORS further and for the requirements of next-generation ocean reanalysis systems.