Abstract
Several satellite-derived water quality and sea surface temperature (SST) products from the Copernicus program are available for coastal environments, but most rely on empirical algorithms that do not fully capture the bio-optical complexity of these waters. Unlike open-ocean conditions, coastal optical properties are shaped not only by chlorophyll-a (Chl-a) but also by total suspended matter (TSM) and colored dissolved organic matter (CDOM), whose variable contributions increase spectral complexity and challenge biophysical parameter retrieval. Consequently, the accuracy and reliability of Copernicus products in coastal zones remain insufficiently assessed and require further investigation. Here, we review and assess Copernicus products for the Northern Adriatic Sea (NAS). Multi-year in-situ measurements from optical and physical sensors are employed to evaluate the quality of satellite-derived products as well as those from biogeochemical models. The accuracy assessment is conducted using several performance metrics, including the coefficient of determination (R²) and the normalized root mean square error (NRMSE). In addition, we propose a bias-correction approach valid for the NAS to further improve the agreement between Copernicus products and in-situ coastal observations. The results show that SST products, whether satellite- or model-based, are highly reliable (NRMSE ≈ 3%). Various Chl-a products also demonstrate promising performance, particularly after bias correction (NRMSE< 10%). In contrast, the turbidity product exhibits reduced reliability, especially for values exceeding 4 FNU. Model-derived Chl-a and salinity products show substantial uncertainties. This study underscores the importance of appropriate bias-correction strategies to ensure the suitability of Copernicus products in optically complex coastal waters.