Fruit growth modeling and realization for greenhouse tomato

Fruit set is the key factor to crop yield and it has been a study focus so far. How to find an indicator that can reflect fruit set mechanism become more and more important. In this paper, utilized greenhouse tomato as a case to study the fruit set mechanism. Tomatoes were planted in solar greenhouse with four densities, environmental data and experiment data were recorded in detailed. Combined with math model, controlled environmental conditions of greenhouse and computer software design technology, the analysis were made to find the related factors which influence the fruit yield and dynamic fruit set rate for different plant density data. GreenLab model had particular advantage to simulate plant growth at organ level. With the help of GreenLab, the dynamic ratios of source to demand (i.e. Q/D) of biomass assimilation were output one growth cycle by one growth cycle. The relationship between the dynamic rate of fruit sets and the dynamic ratio of source to demand (i.e. Q/D) of biomass assimilation was built through the correlation analysis between observed data of dynamic fruit set and calculated Q/D value of model output. In order to computer programming and simulate, tomato topology structure was observed and plant topology generating rhythm was described as List data structure of Scilib language. This data structure can describe main stem and lateral axis alternative growth and syngenesis relationship between organs, so a plant topology structure in time sequence was produced. From the seeds, organs creation, biomass acquisition and partitioning were processed during the same growth cycle to insure feedback between organogenesis and photosynthesis. A global feedback dynamic fruit growth model was successfully built up. Following, independent data was used to validate the model. Both simulation data and measurement data of biomass and geometry size were close. The work provided a new research approach for crop yield. Introducing fruit set into mechanistic models can make growth and development prediction more precisely, especially for fruit. Meanwhile, structure and function variations integration was the highlight of work. This work improved the current GreenLab model as far as fruit growth simulation, and would provide a quantitative tool for the research on fruit sets. Tomato fruit growth model in this research called as GreenLab_Tomato_Fruit model which simulated fruit yield variation with plant density more precisely. The results also demonstrated that season may affect the model parameter, difference densities in seasons needed to further explore in GreenLab_Tomato_Fruit model. Combining with optimization method, the model would provide useful tools to optimize planting density and horticultural practice such as pruning and environmental control for more crops in special constrained environment in the future.