Light interception efficiency of apple trees: A multi-scale computational study based on MAppleT model

The light interception efficiency of a fruit tree plays a key role in its transpiration and photosynthesis, therefore has a strong impact on its growth and yield. Previous research has indicated that manipulations of tree architecture, either through genetics or by agronomic practices, can help in improving light interception efficiency. However, the complexity of fruit tree structure as well as their long growth period makes it difficult to use real trees for exploring the link between architecture and light interception throughout tree development. In order to save time, resource and labour, this study relies on MAppleT, an architectural model of apple tree, and MμSLIM, a multi-scale light interception model part of the Vplants software library, to implement in silico experiments. The influence of apple trees' architecture on their light interception was analysed at four scales: leaves, shoots branches and whole tree. At each scale either the actual position of the radiative components or a turbid medium hypothesis was used to estimate light interception. The STAR, namely the silhouette to total area ratio, was used to evaluate the level of interception efficiency. These investigations aim at comparing light interception estimation on simulated apple trees with bibliographic data and at highlighting STAR variations depending on the underlying assumption, i.e. turbid medium vs. actual position. The results are interpreted in relation with the spatial organisation of components and shoot types within the tree, and its impact on light interception estimation, focusing on deviance from turbid medium assumption.