摘要
树冠是树木进行光合作用和呼吸作用的重要场所,冠幅和冠长率直接影响树木的生活力和生产力。以湖南省黄丰桥国有林场103块杉木(Cunninghamia lanceolata)林样地为例,样地按郁闭度CD≥0.8(类型1)、0.7≤CD<0.8(类型2)、0.6≤CD<0.7(类型3)、0.5≤CD<0.6(类型4)和CD<0.5(类型5)划分5个等级。分析不同等级林分中树木冠幅和冠长率的分布结构以及与胸径的关系,同时利用非线性混合效应模型分析各类型郁闭度对冠幅和冠长率的随机影响。研究结果表明:5种类型的冠幅大小主要分布在区间(2.5—4.0m)之间,分别占65.82%、69.56%、70.79%、58.15%和53.21%,小于1.5 m和大于5.0 m分布较少;5种类型冠长率主要分布在区间(0.3—0.7)之间,分别占55.69%、71.93%、67.01%、82.22%和79.28%,在区间小于0.2和大于0.8分布较少,大于0.9的冠长率几乎没有;冠幅与胸径之间呈现较弱的线性相关性,而冠长率与胸径之间几乎线性不相关。在相同的胸高直径时,类型5的冠幅最大,而类型1的冠幅最小。对于冠长率,恰好相反,即类型1的冠长率最大,而类型5的冠长率最小;同时给出的冠幅模型(模型11)和冠长率模型(模型14)具有较高的预测精度,在实际应用中可以利用它们对冠幅和冠长率进行预测。
Crown size is an important tree variable included in growth and yield models, which are commonly used as decision-support tools in forest management. Crown size is usually described as tree crown width or live crown ratio. Tree crown width and live crown ratio have been found to be a useful measure for tree vigor, and allow one to predict tree growth, mortality and aboveground biomass. In this study, using 103 plots of Cunninghamia lanceolata stand in Huangfengqiao forest farm in Hunan Province as an example, each plot was divided into five classes by stand canopy density （CD） following the rules of CD≥0.8 （Type 1）, 0.7≤CD〈0.8 （Type2）, 0.6≤CD〈0.7（Type3）,0.5≤CD〈0.6（ Type 4）, and CD 〈0.5 （ Type 5 ）. We analyzed the distribution structure of tree crown width and live crown ratio of each tree for each type, and also analyzed the relationships of tree crown width and diameter at breast height （DBH） , live crown ratio and DBH, respectively. In addition, nonlinear mixed effects models approach was used to describe the random effects of canopy density on tree crown width and live crown ratio. The commonly used logistic model with predictors of DBH and mean diameter at breast height in larger trees （MDL） was selected as a based model to construct mixed effects crown width and live crown ratio models. The prediction accuracy of the developed nonlinear mixed effects crown width and live crown ratio models were tested using the validation data. Results showed that： the sizes of tree crown width of five types mainly distributed in the interval from 2.5 （m） to 4.0 （ m）, accounting for 65.82% ,69.56% ,70.79% ,58.15% and 53.21%, respectively, and less distributed in the intervals of less than 1.5 （m） and more than 5.0 （m）. The sizes of live crown ratio of five types mainly distributed in the interval from 0.3 （ m ） to 0.7 （ m ）, accounting for 65.82%, 69.56%, 70. 79%, 58.15% and 53.21%, respectively, and less distributed in the intervals of less than 0.2 （m） and more than 0. 8 （m）, and specially, there is almost no distributed for more then 0.9（m）. The linear correlation between tree crown width and DBH, and live crown ratio and DBH are very weak. The size of crown width of the tree for Type 5 is the largest and is the smallest for Type 1 under the same DBH. For live crown ratio, on contrary, namely the live crown ratio of the tree for Type 1 is the largest and is the smallest for Type 5 under the same DBH. The developed nonlinear mixed effects crown width and live crown ratio models not only described the effects of canopy density on tree crown width and live crown ratio effectively but also have a high prediction. This article main emphasizes on method researches, the effects of many other stand or tree characteristics, such stand age and stand density, on tree crown width or live crown ratio also can be studied using the nonlinear mixed effect models approach.
出处
《生态学报》
CAS
CSCD
北大核心
2013年第8期2434-2443,共10页
Acta Ecologica Sinica
基金
林业公益性行业科研专项(201104028)
国家林业局林业公益性行业科研专项(201204510)
林业公益性行业重点项目(201004002)
关键词
冠幅
冠长率
杉木
非线性混合效应模型
tree crown width
live crown ratio
chinese fir
nonlinear mixed effects model
