Along with the unprecedented availability of satellite data acquisition and technological facilities, monitoring of the Biosphere became priority during last years. At the same time, machine learning (ML) solutions evolved into standard practice to solve prediction problems and speed up processing tasks. With the ambition to overcome limitations related to technical resources in satellite image processing, in this work we implemented the ML algorithm Gaussian process regression (GPR) into Google Earth Engine (GEE) to enable spatiotemporal mapping of vegetation traits at European scale. Also, associated uncertainty is provided, allowing to evaluate robustness of the models. In the case of LAI, deviations lower than 1.2 m2/m2 are observed. The used imagery collection came from the Sentinel-3 (S3) OLCI (Ocean and Land Colour Instrument) top-of-atmosphere (TOA) radiance (L1C) starting from April 2016 until the present date. The generated products were then further used to analyze phenology. A demonstration case is provided over the Iberian peninsula. We observed annual patterns with peaks during spring close to 20 µg/cm2 for LCC (Leaf Clorophyl Content), 1.5 m2/m2 for LAI (Leaf Area Index) and 0.5 for FAPAR (Fraction of Absorbed Photosyntheticaly Active Radiation) and FVC (Fractional Vegetation Cover), calculated as average over the targeted area. Eventually, the developed S3 vegetation products are aimed to support of the FLEX fluorescence mission that is dedicated to monitor vegetation photosynthetic activity.