Abstract
Accurately predicting crop development stage is key to simulating growth and yield formation in crop models. Low temperature stress is a major limitation to global wheat production and can significantly slow down wheat development rate. In a four-year environment-controlled phytotron experiments, detailed phenology datasets were obtained for low temperature stress treatments with different temperature levels and durations. Six widely-used temperature response functions (Linear, Bilinear Triangular, Trapezoidal, Bell-shaped, and Sin) for wheat phenology estimation were combined with WheatGrow model to simulate the low temperature stress effects on elongation-anthesis durations in order to test whether the effects of low temperature stress can be captured by current temperature response functions. In addition, a new algorithm for quantifying daily thermal sensitivity (DTS) was proposed and applied in the six temperature response functions to improve the prediction ability of phenology submodel under low temperature stress. The result indicates that anthesis stage was significantly delayed under low temperature stress at elongation and booting stages. All six original temperature response routines underestimated the delay of wheat development rate caused by low temperature stress, and the Linear and Triangular temperature response routines showed better performance than other four functions. A new improved DTS algorithm (DTSimproved) was proposed which can better quantify the delay of wheat development rate during low temperature stress and the recovery of development rate after low temperature stress. Model validation results show that compared with the DTSoriginal routine, all six improved phenology submodels (DTSimproved) significantly reduced the simulation error under low temperature stress, with the RMSE of elongation-anthesis duration under extreme low temperature stress (Tmin<0) decreased by 53.6% and 36.7% for cv.Yangmai16 and cv.Xumai30, respectively. Therefore, the newly improved routine for wheat phenology under low temperature stress can significantly reduce the uncertainties in model-based impact assessments under low temperature stress.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have