首页|基于竞争与立地效应的湖南栎类天然林冠幅模型

基于竞争与立地效应的湖南栎类天然林冠幅模型

扫码查看
[目的]分析立地因子和竞争因子对冠幅生长的影响,构建含立地因子和竞争因子的混合效应湖南栎类天然林冠幅预测模型,为栎类天然林科学经营决策提供理论依据。[方法]以 51 块湖南栎类天然林样地的1 429株栎类林木为研究对象,筛选对冠幅影响显著的因子,将显著性因子按照标准分级、组合,构成竞争类型和立地类型;从 10 个基础模型中筛选出最优模型;应用k-means聚类方法将初始竞争类型聚类成竞争类型组;将立地类型和竞争类型组作为随机效应加入最优模型,构建含竞争和立地效应的湖南栎类天然林冠幅模型。[结果]对冠幅影响显著的因子包括海拔、坡度、坡位、坡向、土壤类型、相对断面积(RS)、简单竞争指数(CLH)和大于对象木断面积之和(BAL),立地因子显著性顺序为海拔>坡度>坡向>坡位>土壤类型,竞争因子中RS与冠幅呈正相关,而CLH、BAL与冠幅呈负相关;最优基础模型为异速生长模型(有截距),R2 为 0。534 8;将筛选的 3 个竞争因子按照标准分级、组合构成竞争类型,加入基础模型构建含竞争类型的混合效应模型,R2 升至0。583 5;应用k-means聚类将初始竞争类型聚类成17个竞争类型组,作为随机效应加入基础模型,R2 升至0。749 2;将立地类型作为随机效应再加入模型,构建基于竞争和立地混合效应的冠幅模型,R2 升至 0。841 6,相对基础模型提升了57。36%。[结论]含竞争因子和立地因子的湖南栎类天然林冠幅模型具有较好的拟合效果及预测精度。因此,本研究所构建的冠幅模型可以很好地预测湖南栎类天然林的冠幅,可为研究栎类天然林生长、经营管理以及更新森林资源调查数据库提供支持。
Crown diameter model of Hunan Quercus natural forest based on competition and site effect
[Objective]This study was carried out to analyze the effects of site factors and competition factors on the crown width growth,construct a crown width model of Hunan Quercus natural forests with the mixed effects of site types and competition factor.It provides a theoretical basis for scientific management decision of Quercus natural forests.[Method]With 1 429 Quercus trees in 51 natural forest plots of Hunan province as the research object,the factors that had significant influence on crown extent were screened,and the factors were classified and combined according to the standard to form site types and competition types.The optimal basic model was selected from 10 basic models.k-means clustering was used to cluster the initial competition types into competition type groups.The competition type group and the site type group were added into the optimal basic model as random effects,and the mixed effects model including the site type and the competition type group was constructed.[Result]The factors that significantly affected crown width included altitude,slope,slope position,slope direction,soil type,relative fault area(RS),simple competition index(CLH),and greater than the sum of the object wood fault area(BAL).The significance order of site factors was altitude>slope>slope direction>slope position>soil type,and Rs was positively correlated with crown width among competition factors.CLH and BAL were negatively correlated with crown width.The optimal basic model was allometric growth model(with intercept),R2 was 0.534 8.The three competition factors were classified and combined according to the standard to form the competition type,and the mixed effect model with competition type was added to the basic model,and the R2 increased to 0.583 5.k-means clustering was applied to cluster the initial competition types into 17 competition type groups,which were added to the basic model as random effects,and R2 increased to 0.749 2.The combined site types were added to the model as random effects,and the crown width model based on the mixed effects of competition and site was constructed.The R2 of the model increased to 0.841 6,which was 57.36%higher than that of the basic model.[Conclusion]The natural forest canopy model of Hunan Quercus with competition factor and site factor has better fitting effect and prediction accuracy.Therefore,the canopy width model constructed in this study can well predict the canopy width of natural Quercus forest in Hunan,and provide support for the study of the growth and management of natural Quercus forest and the update of forest resource survey database.

crown width modelQuercus natural forestmixed effectsite factorcompetition factor

杨浩、何海梅、黄朗、游文彪、汤维、朱光玉

展开 >

中南林业科技大学 林学院,湖南 长沙 410004

冠幅模型 栎类天然林 混合效应 立地因子 竞争因子

国家自然科学基金项目

32271874

2024

中南林业科技大学学报
中南林业科技大学

中南林业科技大学学报

CSTPCD北大核心
影响因子:1.442
ISSN:1673-923X
年,卷(期):2024.44(6)