Effects of Argo floats data on the ECMWF Ocean DA system and coupled forecasts

The global Argo array plays a pivotal role in ocean observing system by providing nearly uniform global coverage of temperature and salinity profiles to measure the upper 2000 meters of the ocean at approximately a 10-day interval. Argo float data (and other ocean in-situ observation types) are assimilated in the ECMWF ocean and sea-ice analysis system, to provide essential ocean and sea-ice initial states for the coupled forecasting system of ECMWF. In this study, we focus on the impact of ocean observations on medium-range forecasts by taking the global Argo array as an example. Similar studies assessing the impact of atmospheric observations are abundant, while there are few studies examining the impact of ocean observation system. Observation impact on ocean reanalysis has first been evaluated using observation system experiments (OSEs), in which different ocean observation types (including Argo) have been denied in the data assimilation system. Assessment of ocean observation impact on the coupled forecasting system of ECMWF has then been carried out, by initializing the ocean and sea-ice states from different OSE reanalyses. Results suggest that removing Argo data degrades the SST forecasts up to day-10. Impact of removing Argo data is comparable to atmospheric model changes in a typical ECMWF IFS Cycle upgrade and leads to a small but significant degradation of forecasted atmospheric fields.