Impact of interannually variable diffuse attenuation coefficients for downwelling irradiance on biogeochemical modelling

Remotely sensed optical data are fundamental to be integrated into biogeochemical models, since the key role of the optical properties on lower trophic dynamics. In this paper, it is shown how ocean optics satellite products are used to constrain the optical algorithm adopted in the operational biogeochemical model system MedBFM that produces analyses, forecasts and reanalyses for the Mediterranean Sea biogeochemistry within the European Copernicus Marine Environment Monitoring Service (CMEMS).Two different data sets of diffuse attenuation coefficients (Kd) for downward planar irradiance were used in MedBFM to carry out three 15-year simulations. The first simulation is based on climatological values retrieved with a global algorithm; the second uses an updated, interannually variable product from 4 different sensors, obtained with a regional algorithm specifically developed for the Mediterranean Sea; the third is forced by a climatological data set extracted from the updated algorithm. Differences between the two K-d data sets are evaluated in terms of the adopted/differing remote sensing algorithms, and the impact of the two different optical forcings on the MedBFM model output is quantified, with a specific focus on chlorophyll, also distinguishing the effect of using the interannually variable K-d by the one related to the algorithm upgrade.The differences between the interannually variable K-d data set and the climatological one amount to 10%, resulting in local variations of chlorophyll vertical profile concentration, larger than 20% in some periods of the year. Noticeable effects are also observed on the along-basin zonal range of deep chlorophyll maximum depths during the simulated period, which increases with the use of the updated K-d data set. In the western Mediterranean, interannual variability of chlorophyll in summer grows up to 40% at 100 m.Matching-up the updated model outputs with quality-controlled Biogeochemical-Argo floats data of fluorescence-derived chlorophyll results in a small increase of the model skill.