Synthesis of the performance of the integrated modelling systems and recommendations on all 25 products for use by the Copernicus Marine Service (9.1)

The NECCTON project has developed an integrated modelling system to deliver new variables and products to the Copernicus Marine Service. The system is based on the models of the lower trophic levels (LTL) of the marine ecosystem which are already used in Copernicus Marine Service production, coupled to new models that increase the range of available variables to include marine species and stressors. NECCTON has created a set of multi-decade model runs to demonstrate the capability of the modelling system and the quality and technological readiness level of the 27 thematic variables (originally 25 products) proposed. This report summarises the validation and recommendations for each variable produced in the NECCTON hindcasts. The datasets themselves, with validation information, can be found on the NECCTON data portal https://data.neccton.eu. Each regional dataset has been assigned a technology readiness level (TRL) ranging from 4 (tested in a laboratory environment – the model has run successfully and produced a dataset) to 5 (validated in a relevant environment – the dataset has an acceptable match to observations). NECCTON will apply many of the hindcast outputs in the case studies now in progress, and successful completion of a case study will move them to TRL 6 (technology demonstrated in relevant environment). The following thematic variables have reached TRL 5 for at least one region and are recommended for consideration by the Copernicus Marine Service: • LTL: mesozooplankton, particulate organic matter, reflectance • Benthic: bottom oxygen, pH and light, macrozoobenthos and benthic flora • HTL: small pelagics, large pelagics, demersal fish and unspecified fish biomass, ideally using an ensemble of models • Pollutants and stressors: plastics, organophosphate esters, mercury, fisheries pressure and climate change stressor indices. Further recommendations include testing these variables in a wider range of regions and taking steps to address the shortage of observational data.