Relationships between Tomato Cluster Growth Indices and Cumulative Environmental Factors during Greenhouse Cultivation

Abstract

Growth characteristics of plants are mainly defined by genetic information and affected by environmental factors, such as air temperature, solar radiation, and humidity. There is a strong potential and demand to predict plant growth by environmental factors. Therefore, this study established mathematical models via multiple linear regression analysis (MLR) to describe the relationship between tomato cluster's growth indices and environmental factors in greenhouse cultivation. The Japanese tomato variety (Solanum lycopersicum L. Var. Culta. Momotarou Haruka) was cultivated in a hydroponic culture system using a porous solid medium from September 2019 to June 2020 in a glass greenhouse. The significant correlations among growth indices and environmental factors showed that the cumulative solar radiation (CSR) had a stronger effect on the number of flowers (NFl) rather than the cumulative heat unit (CHU) and vapor pressure deficit (VPD). Meanwhile, the number of fruits (NFr) relied much on VPD than CHU and CSR. Also, pollination condition was sensitive to VPD; NFr and enlargement of fruit cells during fruit maturation might be important factors in fruit-cluster weight (CWt). Mathematical models via MLR of cumulative environmental factors, such as CHU, CSR, and VPD could explain tomato cluster's growth indices, such as NFl, NFr, fruit perimeter (PFr), CWt, and duration of tomato clusters with the coefficient of determination (R2) ranging from 0.742 in the fruit maturation duration (FM) model to 0.953 in the PFr model. In MLR models, CHU was the most important factor in the duration of tomato clusters and PFr with the relative importance metrics (RIM) of ranging 29.47%–43.66%; CSR for NFl with RIM was 36.83%; and VPD for NFr and CWt with RIM were 37.37% and 29.37%, respectively. By MLR analysis, results showed the potential of using CHU, CSR, and VPD to describe and predict the growth indices of tomato clusters by mathematical models.