Scientific workflow for reusing plant/FSPM models

For many years a large collection of models have been developed to describe plants (e.g., trees, crops). However, few of them can be reused directly to be integrated into more complex models or combined with other plant models to answer scientific questions. This leads to a loss of time spent in re-implementing published models before starting new projects. To tackle this challenge, a platform called OpenAlea has been developed over the last 10 years with the aim of sharing models and tools among the OpenAlea community. This platform uses the well-known Python programming language to link heterogeneous components/models together. It performs the evaluation of complex models expressed as a dataflow where each node is a different piece of the model. A visual front-end, VisuAlea, allows users to assemble different components together to answer practical questions without writing a single line of code. In this article, we propose to demonstrate the functioning of the platform with a single model/component. As a first step, this component will be connected to statistical components to demonstrate the ability of the platform to share common analysis and plotting functions. In a second step, the same component, instead of taking fixed parameters, will be attached to external models that will provide estimated values for these parameters, showing the ability of the platform to connect models developed by different teams. The third step will be to connect a sensitivity analysis component to this system. Finally, we will propose a different evaluation of the component as part of a temporal simulation to demonstrate how a simple static model can be reused to construct complex plant growth models. (Résumé d'auteur