Rötzer T, Grote R, Pretzsch H
Technical University Munich, Department of Ecosystem- and Landscape management, Chair of Forest Yield Science, Am Hochanger 13, 85 354 Freising, Germany.
Int J Biometeorol. 2004 Feb;48(3):109-18. doi: 10.1007/s00484-003-0191-1. Epub 2003 Oct 15.
A phenology model for estimating the timings of bud burst--one of the most influential phenological phases for the simulation of tree growth--is presented in this study. The model calculates the timings of the leafing of beech (Fagus sylvatica L.) and oak (Quercus robur L.) and the May shoot of Norway spruce (Picea abies L.) and Scots pine (Pinus sylvestris L.) on the basis of the daily maximum temperature. The data for parameterisation and validation of the model have been taken from 40 climate and 120 phenological stations in southern Germany with time series for temperature and bud burst of up to 30 years. The validation of the phenology module by means of an independent data set showed correlation coefficients for comparisons between observed and simulated values of 54% (beech), 55% (oak), 59% (spruce) and 56% (pine) with mean absolute errors varying from 4.4 days (spruce) to 5.0 days (pine). These results correspond well with the results of other--often more complex--phenology models. After the phenology module had been implemented in the tree-growth model BALANCE, the growth of a mixed forest stand with the former static and the new dynamic timings for the bud burst was simulated. The results of the two simulation runs showed that phenology has to be taken into account when simulating forest growth, particularly in mixed stands.
本研究提出了一种用于估算芽萌动时间的物候模型，芽萌动是树木生长模拟中最具影响力的物候阶段之一。该模型根据日最高温度计算山毛榉（欧洲山毛榉）、橡树（夏栎）、挪威云杉（欧洲云杉）和苏格兰松（欧洲赤松）的展叶时间以及挪威云杉的五月新梢时间。模型参数化和验证的数据取自德国南部的40个气候站和120个物候站，温度和芽萌动的时间序列长达30年。通过独立数据集对物候模块进行验证，结果表明，观测值与模拟值比较的相关系数分别为：山毛榉54%、橡树55%、云杉59%、松树56%，平均绝对误差在4.4天（云杉）至5.0天（松树）之间。这些结果与其他（通常更复杂的）物候模型的结果非常吻合。在将物候模块应用于树木生长模型BALANCE后，模拟了具有先前静态和新动态芽萌动时间的混交林分的生长情况。两次模拟运行的结果表明，在模拟森林生长时，尤其是在混交林中，必须考虑物候因素。
