Towards continuous, evidence-based evolution of Copernicus Marine Service through open science practices : insights from the IMMERSE project.

The Copernicus Marine Service relies in a very fundamental way on digital tools whose deployment and routine implementation allow the transformation of information from observation networks into products available to users. In practice, these digital tools are software whose development is partly carried out in the framework of interaction with the academic community. These software packages bring together and summarize our collective understanding of ocean and sea-ice dynamics,  observation networks and inversion techniques. 

The evolution of the Copernicus Marine Service therefore necessarily involves the evolution of these software tools. A significant part of our community devotes energy and time to improving these software and the processes enabling its deployment in operational systems. The development of these software tools is therefore a central question in the transition from research to operation within the Copernicus Marine Service. 

The NEMO modelling framework is an example of such scientific software on which the Copernicus Marine Service is based. This codebase is used as a tool to understand ocean and sea ice dynamics, to prepare observational networks, and to integrate information from observing systems into reanalysis and forecasts. In this it concentrates the work of a very large community on a very long time scale. Its evolution is in any case key for the evolution of the Copernicus Marine Service and its downstream applications. 

Over the last four years, the consortium of the IMMERSE project has accompanied the development of the NEMO code and the organization of the transition of this development within the Copernicus Marine production centers. The project activities have led to a significant improvement in the computational efficiency of the code, to a better representation of the key physical processes. It also allowed the exploration of development practices based more fundamentally on open science approaches. 

In this presentation we will propose a synthesis and a feedback on the contribution of these approaches to open science and on the stakes of the transition from research to operation in the framework of Copernicus Marine Service. We will recall the main results of the IMMERSE project and in particular the actions taken to accelerate this transition. We will also use the example of the collaborative data challenges deployed in the framework of altimetry data processing. We will try to paint an objective picture of the contribution of new collaborative work practices in the field.