Using critical points of logistic model to describe the growth of rice plant height in Taiwan

The height of rice plants is not only an important component of the crop canopy structure but also a crucial pathway for increasing crop yield. In this study, we used a logistic regression model to fit the plant height data of rice varieties cultivated in Taiwan over years and interpret the parameters from the perspective of crop growth. The logistic model has five critical points which allows for the inference of various growth stages: absolute acceleration point (AAP, completion of seedling establishment), maximum acceleration point (MAP, tillering initiation), inflection point (IP, effective tillering), maximum deceleration point (MDP, panicle initiation), and asymptotic deceleration point (ADP, heading). We found that the autumn cropping season reached the point of maximum growth rate earlier (AAP: 5; MAP: 11; IP: 19; MDP: 28; ADP: 34 days) than the spring cropping season, with noticeable advancements in the critical points of IP, MDP, and ADP. According to the model parameters, the period between AAP and ADP is the main growth stage of rice plant height, with the maximum growth rate of autumn crops exceeding that of spring crops. The results showed that there was no significant difference between the early and recent varieties in terms of the time to reach the maximum growth rate and its slope for both cropping seasons. The model can be applied to rice cultivation management to schedule the timing of fertilizer application and irrigation.