Assessment of static ensembles on a global ocean assimilation system using EAKF

The establishment of a global ocean assimilation system is of great significance for the utilization of observational data. In this study, a new global ocean assimilation system was established using the EAKF assimilation module and the MPI-OM global ocean circulation model with a grid refining the South China Sea (SCS) and part of Pacific region. Considering computational cost, static ensemble was utilized as the background ensemble. To assess the role of static ensembles, three different static ensembles were selected to compare assimilation experiments. The assimilation of sea surface temperature (SST) and sea surface salinity (SSS), as well as Argo profiling data, was conducted. Firstly, the model performance of the assimilation system on a global scale was evaluated using the profile data, which can reduce the temperature and salinity biases by approximately 50%, resulting in a decrease of the globally-averaged temperature and salinity biases to 0.4 ℃ and 0.1 PSU, respectively. Considering about the specific of the grid, we focus on the SCS and the equatorial region. Comparison of buoy data with the model results indicates that assimilation significantly improves the simulation of the tropical thermocline, with correlation coefficients suggesting that the assimilation with seasonal static ensemble yields the best results. Secondly, the impact of different ensembles on the SCS where significant monsoon influence exists was investigated, revealing distinct differences in current field among the three assimilation experiments. For the experiments with background ensembles include information on seasonal variations, the assimilation improves the model’s ability in reproducing the intrusion of the Kuroshio Current into the SCS.