Simulation Model for Predicting Fruit Yield of Tomatoes Grown on a Single-truss System under Shade in Summer

抄録

This study was carried out to make a simulation model for predicting the fruit yield of tomatoes in summer. Tomato seeds were sown monthly from February to September. Plants were grown hydroponically on a single-truss system in 0, 30, 55, and 83% shade. Total fruit yield per plant increased linearly with increasing integrated solar radiation from the first flower anthesis to the first fruit harvest. When all data from all crops were pooled, the regression line did not have as high a coefficient of determination (r²=0.73) as it did with the unpooled data. Most of the regression lines fitted to the data from each crop individually had higher coefficients of determination, except for lines plotted for the June and July crops. The slope (r²=0.929) and the x intercept (r²=0.860) of the regression lines, except for the June crop, were correlated with averaged air temperature during fruit development linearly and quadratically, respectively. From these results, the simulation model for predicting fruit yield from both integrated solar radiation and averaged air temperature was constructed. The model could be applied at 19 to 27°C of averaged air temperature. Standard error of cross-validation (SECV) of the data excluding the June crop was lower (92.1) in the model modified with air temperature, compared with the model with only integrated solar radiation (111.4).