Assessing the Potential Impact of Changes to the Argo and Moored Buoy Arrays in an Operational Ocean Analysis System

摘要

A series of observing system simulation experiments (OSSEs) have been carried out using the Met Office global Forecasting Ocean Assimilation Model (FOAM) to provide insights on the current and future design of the in situ observing network for ocean monitoring and forecasting. Synthetic observations are generated from a Nature Run (NR) that represents the true ocean state in the experiments. These observations are assimilated in FOAM and the results are compared to the NR to assess the impact of the observations, as well as assessing the effectiveness of the data assimilation system. The NR and FOAM based OSSEs have different resolutions and are driven by different surface forcing. The results show that assimilating observations equivalent to the current observing system allows the system to produce a realistic representation of the ocean state. Additional Argo profiles in some of the Western Boundary Current (WBC) regions and along the Equator improve the performance of FOAM by reducing the root mean square error (RMSE) against the Nature Run by ~10% for temperature and salinity fields in the upper ocean. Assimilating additional Deep Argo floats leads to ~20% RMSE reduction in basin scale regions and the reduction rate is up to 80% in the Labrador Sea below 2500m. An experiment withdrawing mooring profiles indicates the impact of moorings is localised and on average the analysis shows ~5% degradation without the mooring observations. The additional Argo profiles in the WBC regions and deep ocean also have impacts on the representation of the Ocean Heat Content (OHC) and the Atlantic Meridional Overturning Circulation (AMOC), with the deep Argo observations correcting the model drift in OHC below 2000m. The results highlight the necessity of a well-designed and coordinated in situ observing network globally, as well as requirements for future model and assimilation developments to achieve the best use of the additional in situ observations.