The Marine Environment Reanalyses Evaluation Project MER-EP, towards an improved knowledge of the global ocean environment of the past decades, to support ocean reporting and ocean prediction.

Ocean reanalyses are reconstructed past ocean states by combining ocean numerical models and observations through data assimilation techniques. They are key in the seamless ocean information value chain. Thanks to their temporal and spatial consistency, continuity, and high accuracy, ocean reanalyses are widely used for ocean and climate studies in the academic and private sectors.  

The MER-EP project we present here is building on previous ocean reanalysis intercomparison exercises such as the Ocean Reanalyses Intercomparison Project (ORA-IP), and on the joint efforts of the ocean prediction community (Copernicus Marine Service, Oceanpredict/ForeSea/OP-DCC), the ocean and climate modelling research community (CLIVAR/GSOP), and on the Ocean Physics and Climate panel of the  Global Ocean Observing System (GOOS/OOPC) research programme.  

Previous intercomparison exercises of ocean reanalyses have targeted specific variables, such as ocean heat content, mixed layer depth, and ocean transport, to assess the consistency and discrepancies among various ocean reanalysis products.  Future intercomparisons should complement this approach including more systematic regional focus and evaluating different ocean reanalyses to determine their quality and fitness-for-purpose for specific applications and whenever possible provide insights on their conditions of use (eg. which type of reanalysis with which resolution or resolved processes for which application, in which region, etc…).  

Ocean reanalysis producers, relevant scientific programmes  and users are willing to get involved in a new evaluation and intercomparison exercise including new variables, new point-of-views (user oriented) and new focus areas, with studies focused on relevant use cases.  

In this context, the primary objective of MER-EP is to improve our knowledge of the ocean by understanding and ultimately improving the reliability and usability of global and regional ocean reanalyses, including physics, waves, biogeochemistry, and sea ice, based on representative and high-priority use cases identified after extensive discussions with academic and private sectors ocean reanalyses users. The corollary objective is to improve our knowledge of the capabilities of similar forecasting systems and of emerging AI forecasting systems trained using these reanalysis datasets, at the global and regional scales, and for a range of applications.