AbstractIn this work, we couple the functional-structural plant model CPlantBox to the Unreal Engine by exploiting the implemented raytracing pipeline to evaluate light influx on the plant surface. There are many approaches for photosynthesis computation and light evaluation, though they typically are limited by versatility, compute speed, or operate on much coarser resolutions. This work specifically addresses the concern that data generation pipelines tend to be unresponsive and do not include model-based knowledge as part of the generation pipeline. Using established photosynthesis solvers, we model the interaction between the Unreal Engine and the FSPM to measure physical properties in the virtual world. This is successful if we are able to reproduce experimental results using an in silico model. As part of the pipeline, we generate a surface geometry and utilize material shaders that are designed to establish a baseline surface model for light interception and transmission, based on simple parameter sets that can be calibrated. Using a Selhausen field experiment as baseline, we reproduce the photosynthesis effectiveness of the plants in the 2016 winter wheat experiments. Our pipeline is deeply intertwined with data generation and has been proven to perform well at scale. In this work, we build on our previous work by showcasing both a simulation study of a light evaluation as well as quantifying how well our system performs on high-performance computing systems.Competing Interest StatementThe authors have declared no competing interest.Footnotes* https://github.com/dhelmrich/Synavis* https://doi.org/10.6084/m9.figshare.28280780.v2...

Virtual World Coupling with Photosynthesis Evaluation for Synthetic Data Production

10.1101/2025.02.06.633870