Evaluation of a radiosity based light model for greenhouse cucumber canopies

Light distribution is a key factor of developmental and growth processes, and strongly depends on the foliage distribution which is affected, e.g., by the arrangement of the plants in the canopy. The precise simulation of the light distribution on organ level is an essential component for dynamical plant models which incorporate structural and physiological adaptions of plants to their environment. Combinations of static 3D plant models with 3D light models are used for analyzing the complex light distribution on leaf level in silico, but detailed measurements for evaluation of simulation results are almost non-existent. This study addressed the evaluation of a model on a high level of detail using individual leaf based light measurements in canopies of cucumber (Cucumis sativus L.). We combined a static 3D plant model derived from digitized plants on an individual organ scale with a mock-up of the surrounding canopy and a 3D radiosity based light distribution model. Variations of plant density and spacing were analyzed to cover a range of canopy architectures. An exclusion of components of the light environment by applying a shading encasement followed by a successive uncovering allowed investigating the scene under increasing levels of complexity. The combined 3D plant-light distribution approach allowed determining the interaction of the light directions and the canopy architecture as well as differences in the accuracy of the simulations. Depending on canopy architecture and shading treatment, the light distributions covered a range from exponentially shaped vertical gradients in encased treatments to nearly flat light profiles in nonencased conditions. In conclusion, simulations of leaf level PAR based on combinations of detailed 3D surfaced-based plant and light distribution models are suitable to derive light-induced physiological responses on organ level.