FFAST aims at describing wheat genotypes functioning through an innovative model-assisted phenotyping strategy. Currently, studies on field phenotyping are mostly focused on exploiting directly structural traits observations (e.g. leaf area, height) to establish statistical models with genetic characteristics. However, structural traits are highly determined by the environment, and such empirical models are insufficient to describe genotypes functioning. FFAST proposes an alternative approach using functional plant models (FPM, also known as crop process-based models) to describe the eco-physiological mechanisms that produce a differentiated response of the genotype to the environment (GxE). This model-assisted strategy consists in assimilating large observational datasets of multiple structural traits over different growing environments to retrieve, for each genotype, a set of varietal parameters of a FPM. These varietal parameters describe the genotype functioning and constitute functional traits, closely linked to its genetic characteristics. The model-assisted phenotyping method will be evaluated in a panel of ten bread wheat genotypes that will be monitored on phenotyping experiments and by satellite. The phenotyping experiments will be conducted in the Toulouse, Clermont-Ferrand and Montpellier sites –part of the PHENOME-EMPHASIS phenotyping infrastructure– during three years. That will permit to acquire high-throughput observations of multiple structural traits (leaf area, canopy height, heading date, ears density…) in different environments. Nevertheless, as a large environmental variability is essential to retrieve accurately functional traits, FFAST will investigate the use of high-resolution satellite platforms to provide additional cost-efficient observations of structural traits for specific genotypes over contrasted environments. Three genotypes of the panel will be monitored by satellite on 40 distant commercial fields over a climatic gradient in eastern France. Images from Sentinel 2 and PlanetScope satellite constellations will be used to retrieve frequent observations of some essential traits like the leaf green area index (GAI) and the fraction of absorbed photosynthetically active radiation (fAPAR). The estimation of functional traits from the observations will rely on a data assimilation framework based on the Sirius Quality FPM, specifically developed for wheat, which will be linked to the architectural model Adel Wheat. This will permit to improve the description of structure-driven processes such as light interception/absorption or evapotranspiration. Bayesian Monte Carlo methods will be used to retrieve varietal parameters of Sirius Quality from the structural traits observations for each genotype. The resulting posterior distribution of varietal parameters for all the genotypes will be analysed to identify those parameters –or groups of parameters characterizing the same mechanism– presenting statistically different posterior distributions among genotypes. Those parameters will constitute functional traits. The approach proposed by FFAST will be validated evaluating the reliability of the functional traits identified to predict the genotype performance in different environments from those used during the assimilation. This will permit to evaluate as well the role of remote sensing observations over different environments in the FFAST approach, compared to expensive multi-site phenotyping experiments. The project results will be disseminated through scientific papers in different domains: phenomics, eco- physiology, crop modelling and remote sensing. The observational datasets collected for the 10 genotypes will be also made public through a data paper. Moreover, the development of a methodology to produce multi-constellation GAI and fAPAR observations suitable for plant phenotyping will permit HIPHEN –enterprise partner in FFAST– to open new commercial services.