Abstract:To quantify dry matter partitioning among different fruit trusses and to improve production efficiency of greenhouse tomato, three experiments were carried out for over-winter, early-spring and spring-summer crops, respectively. Total dry weights of stem, leaves, fruits within a source-sink growth unit (one fruit truss and three leaves just below the truss) were measured during growing periods. A simulation model of total biomass and the partitioning among stems, leaves and fruits within a growth unit was constructed by combining a classical model of dry matter production in a leaf with the GreenLab model of dry matter partitioning. Our combined model was tested against measured data. The slope and coefficient of determination (R2 ) between simulated and measured biomass of the whole plant for three experiments were 0.93 and 092, respectively. The slopes between simulated and measured dry weight of stems, leaves and fruits within source-sink growth units and dry weight of root of the plant were 0.85-0.89, and the relative error between them was 5.3%, 5.6%, 8.1% and 3.6%, respectively. These results suggested that our model could be used for optimizing fruit yield of greenhouse tomato production.