Résumé
By formalizing traits as the result of genotypic and environmental effects and the relations among traits, ecophysiological models (or process-based models) provide a platform for integrative analyses of trait impacts on whole-plant and crop phenotypes. Over the past two decades, model development has been increasingly driven by the need to account for genotypic differences across environments, and improvements in this area have developed around well-defined traits such as leaf elongation, early vigor and flowering time. Process-based models are now increasingly used to define and characterize crop environments at various scales and help breeding programs take advantage of G × E interactions. Ecophysiological models are also used to assist plant phenotyping and could provide necessary links between controlled-conditions phenotyping and plant performance in the field. The integration of genetic controls in ecophysiological models has allowed analysis of the genetic control of phenotypic plasticity across wide ranges of environments, and the G × E × M space is now explored using efficient algorithms to find ideotypes optimizing many antagonist criteria. This later approach lies in finding combinations of values of the genetic and agronomic parameters that best satisfy the pre-defined objectives, but it is currently limited by the lack of quantitative relationships between genes and model parameters. Considerable efforts are still needed to develop robust links between genetic controls, physiological determinants and traits relevant to breeders.