To quantitatively analyze the regularity and characteristics of tillering dynamics of rice at different sowing dates. In this paper, the whole process of rice tillering was decomposed into two aspects: tiller occurrence and tiller extinction, and two Logistic functions were used to describe them respectively, so as to establish a dynamic tillering model in rice based on double Logistic function (Dynamic Tillering Model, DTM). Then, according to DTM, dynamic tillering indexes (DTIs) were defined and their calculation formulas were derived. Finally, the characteristics and laws of rice tillering dynamic response to sowing dates were analyzed using DTM and DTIs with the observed tillering data of three rice varieties, (Hui-Liangyou 898 (HLY898), Y-Liangyou 900 (YLY900) and Y-Liangyou 911 (YLY911)), in six sowing dates (March 15, March 20, March 25, March 30, April 5, April 10). The results show that: 1. The model fits well. The normalized RMSE (RMSEn) of the DTM fitted to the observed tillering data of different rice varieties sown at different times were all less than 10 %, and their mean values were less than 5 %. 2.The variation degree of DTIs under the influence of sowing dates had certain consistency among the three rice varieties. the inherent rate of tillering (Rit), the maximum tillering rate (Rmt), the maximum tillers extinction rate (Rme) and the duration of tillering (Dt) varied greatly, while the total number of grow tillers (Ng), the peak time of tillering (Tpt), the peak time of tillers extinction (Tpe) and the end time of tillering (Tet) had smaller variation. 3. The eight DTIs, the inherent rate of tillering (Rit), the duration of tillering (Dt), the maximum tillering rate (Rmt), the number of retained tillers (Nr), the peak time of tillering (Tpt), the end time of tillering (Tet), the start time of tillers extinction (Tse) and duration of tillers extinction (De), had a consistent linear response to the sowing dates among the three rice varieties. 4. Under different sowing dates, the dynamic characteristics of tillering of YLY900 and YLY911 were relatively close, while HLY898 had great differences from YLY900 and YLY911. In this paper, the evolution process of the number of tillers of rice was accurately described by the DTM, and the regularity and characteristics of tillering dynamics of rice were deeply revealed using the DTIs, with agronomic experiment of three varieties with six sowing dates. Therefore, it has important theoretical value to deeply understand the law of the tillering dynamic of rice affected by sowing dates and has important practical significance for guiding accurate planting and fine management of rice production from the perspective of grasping tillering dynamics.
Read full abstract