Analysis of prediction methods for budburst days based on the phenological observation in 29 broad-leaved tree species for 10 years
2007
Fujimoto, S.(Shizuoka Univ. (Japan). Faculty of Agriculture)
Budburst phenology was observed in 29 broad-leaved tree species for 10 years at the Kamiatago Field of Shizuoka University, Japan, located in the warm temperate zone. Prediction methods for budburst days were developed using CT (Cumulative Temperature) and DTS (the number of Days Transformed to Standard temperature) models. In both models, the frequency distribution of starting days of accumulation minimizing the prediction error indicated the lowest value in the case of 45 days from January 1. In the CT model, the threshold temperature values minimizing the standard error were approximately from -7.5 to -2.5 deg C , while in the DTS model, the Ea values (temperature characteristic) minimizing the error were approximately between 10 and 15. Thus, these parameter sets consisted of starting days and threshold temperatures (or DTS) minimizing the prediction errors which enable the realization of adequate methods to predict budburst days. In discussion, applicability and problems in the two models were mainly discussed.
Показать больше [+] Меньше [-]