油松(Pinus tabulaeformis)树轮宽度与气候因子统计相关的生理机制——以贺兰山地区为例

A case study of physiological characteristics of statistical correlation between Pinus tabulaeformis tree-ring widths and climatic factors

通过分析贺兰山地区气候因子的变化特征,并且参考北美、俄罗斯的研究结果,确定了Vaganov-Shashkin生理响应模型的参数,并根据研究区的实际情况对参数作了相应修正,得出两组不同的参数,两组参数拟合出相同的结果,说明模型内在的机理是稳定的,更能反映出它的生理学特性.对树轮宽度序列和气候因子进行相关分析,发现比较明显的相关有与5～8月份降水的正相关,与5～8月份温度的负相关,与9、10月份温度的正相关.模型分析结果认为,从5月中旬到8月底,降水不能满足树木生长的需要,故形成树轮宽度与该时段降水的正相关,而在这个时段,由于温度和降水呈负相关,表现为树轮宽度与温度的负相关,其实这在很大程度上是一种间接的相关.到9月份,由于温度下降比较明显,这时树木生长的主要控制因素已转化为温度,所以形成了与9、10月份温度的正相关.模型分析还发现,树轮宽度的形成与生长结束时间相关性强,而与生长开始时间的关系不明显.

The Mt. Helan is located in semi-arid area of North China. It is a typical region for dendrochronology study, to date, there are lots of climatic reconstructions for this region, but very few physiological response analysis though it is very important. In this paper, tree growth physiological responses to climatic factors are discussed by using the Vaganov-Shashkin model for the Mt. Helan region. The results could explain the relationship between tree-ring widths and climatic factors more reasonably. The materials used in the paper include meteorological data from three meteorological stations and one tree-ring width series. Considering the small difference of the elevation between AZQ meteorological station and tree ring site, the meteorological data are calibrated by a method described following： minus 0.9℃ from each datum of temperatures, and times 1.2 for each of precipitation data. In spite of that the parameters in Vaganov-Shashkin model have been fixed based on the research from the North America and Russia, they have been adjusted in terms of the factual situation of the research region. We therefore obtained two groups of parameters, which were applied in the V-S model separately, and the simulation results are quite similar. This indicates that the model is stable and has clearly physiological characteristics. There are certain correlations between tree-ring width series and climatic factors： positive with the precipitation from May to August; negative with the temperature from May to August, and positive with September to October. The simulation results indicate that the precipitation from the second ten days of May to August is insufficient to tree＇s growth. With the decreasing of temperature in September and 6ctober, the main controlling factor turns from precipitation to temperature. We also, in this paper, found a strong relationship between ring widths and the ending of growing season, and weak correlation with the beginning of growing season.