Correlation Analysis for Strongly Coupled Atmosphere-Ocean Data Assimilation for Future Versions of the Global Forecast System

The Joint Effort for Data assimilation Integration (JEDI) framework aims to unify data assimilation (DA) programs across modeling systems. The upcoming version of the National Oceanic and Atmospheric Administration’s (NOAA) Global Forecast System (GFS) is planned to be coupled (atmosphere, ocean, sea-ice, waves) with each initialization performed individually for model components. This is known as weak coupling. Using the JEDI framework, one can test strong coupling, in which the model components' DA are initialized together to determine the importance of the implementation of strong coupling in future versions of the GFS. This project analyzes the strongly coupled ocean-atmosphere data assimilation between the Finite­-Volume Cubed-Sphere Dynamical Core atmospheric model (FV3) and the Modular Ocean Model 6 (MOM6). To evaluate this coupling, high-resolution sea surface temperature (SST) from the MOM6 and 2-meter air temperature (T2m) from the FV3 were compared via difference and correlation analysis with daily data from eleven days in the summer and winter. Correlations are seen to change temporally and spatially. T2m–SST correlations over time showed the strongest relationships in the summer hemisphere and certain specific locations such as the El Nino Southern Oscillation (ENSO) region. Correlations were weakest towards the poles with increasing sea ice and colder temperatures. Difference plots (SST-T2m) also demonstrated spatial differences between the two components and showed significant flow dependence. These results aid model developers in determining where and when strongly coupled data assimilation is effective.