首页|塔里木河下游不同径级胡杨死亡的影响因素

塔里木河下游不同径级胡杨死亡的影响因素

扫码查看
树木死亡威胁森林的正常更新和演替过程,导致不同径级树木死亡的影响因素是否具有一致性,目前尚不清.以塔里木河下游荒漠河岸胡杨林为研究对象,结合生物因素(树木胸径、同种邻体密度)和非生物因素(土壤盐分含量、pH值、含水率等),通过逻辑斯蒂广义线性混合模型和随机森林模型,分析不同径级胡杨死亡的驱动因素及其相对重要性.结果表明:对于胸径≤10 cm的胡杨个体而言,土壤盐分含量对胡杨死亡的影响最大,胡杨死亡率随着土壤盐分含量的升高而上升,影响其次的是土壤含水率和铵态氮含量,而半径20m同种邻体密度和胸径的影响最小.对于10 cm<胸径≤30 cm的胡杨个体而言,胸径对胡杨死亡的影响最大,胸径的增大极显著降低了胡杨死亡率(P<0.001),影响其次的是土壤盐分含量和半径 30m同种邻体密度,土壤全磷含量和pH值的影响最小.对于胸径>30 cm的胡杨个体,半径 15m同种邻体密度对胡杨死亡的影响最大,胸径和土壤盐分含量的影响其次,土壤铵态氮、有机碳含量的影响最小.随机森林模型的预测准确率较高,能够有效分类和预测大部分胡杨个体死亡情况,可为塔里木河下游天然胡杨林可持续管理以及掌握胡杨死亡机制提供理论指导.
Influencing factors of the mortality of Populus euphratica trees in different diameter at breast height classes in the lower reaches of the Tarim River
Tree death severely threatens the regular regeneration and succession processes of forests.However,it remains unclear whether the factors influencing the mortality of trees with different diameters at breast height(DBH)are consistent.This study explored the factors influencing the mortality of the Populus euphratica desert riparian forest in the lower reaches of the Tarim River.The biotic factors(e.g.,DBH and conspecific neighbor density)and abiotic factors(e.g.,soil salinity,pH,and moisture content)were incorporated into logistic generalized linear mixed models and random forest models to predict the driving factors and their relative importance in the mortality of Populus euphratica plants in different DBH classes.The results indicated that for the individuals with DBH≤10 cm,soil salinity had the greatest impact on mortality,and the mortality increased significantly with the increase in soil salinity.Soil moisture content and ammonium nitrogen content were also influential,whereas conspecific neighbor density within a 20 m radius and DBH had minimal effects.For the individuals with 10 cm<DBH≤30 cm,DBH was the most significant factor,and larger DBH reduced the mortality(P<0.001).Soil salinity and conspecific neighbor density within a 30 m radius also played important roles,while total phosphorus content and soil pH had minimal effects.For the individuals with DBH>30 cm,conspecific neighbor density within a 15 m radius exerted the greatest influence,followed by DBH and soil salinity,while ammonium nitrogen and soil organic carbon had the least impact.The random forest model demonstrated high predictive accuracy,effectively classifying most Populus euphratica individuals and predicting the mortality.The findings provide theoretical guidance for the sustainable management of Populus euphratica riparian forests in the lower reaches of the Tarim River and contribute to understanding the mechanisms driving the mortality of Populus euphratica.

Tarim RiverPopulus euphraticatree deathgeneralized linear modelrandom forest

杨文宇、玉米提·哈力克、马映栋、何佳宁、艾力甫达尔·阿尔肯、陈凯旋

展开 >

新疆大学生态与环境学院,新疆 乌鲁木齐 830017

绿洲生态教育部重点实验室,新疆 乌鲁木齐 830017

塔里木河 胡杨 树木死亡 广义线性模型 随机森林

2025

森林与环境学报
福建农林大学

森林与环境学报

北大核心
影响因子:0.964
ISSN:2096-0018
年,卷(期):2025.45(2)