欢迎访问林业科学,今天是

林业科学 ›› 2026, Vol. 62 ›› Issue (11): 34-48.doi: 10.11707/j.1001-7488.LYKX20260123

• 前沿热点 • 上一篇    下一篇

气候与竞争对长白落叶松枝条生长的影响及预测模型

苗铮,赵雪晗,蒋宇萌,董利虎*()   

  1. 东北林业大学林学院 森林生态系统可持续经营教育部重点实验室 哈尔滨150040
  • 收稿日期:2026-03-02 修回日期:2026-04-20 出版日期:2026-11-10 发布日期:2026-11-20
  • 通讯作者: 董利虎 E-mail:donglihu2006@163.com
  • 基金资助:
    国家自然科学基金青年科学基金项目(C类)(32301581);国家重点研发计划青年科学家项目(2022YFD2201800)。

Predictive Modeling of Branch Growth of Larix olgensis under Climate and Competition Effects

Zheng Miao,Xuehan Zhao,Yumeng Jiang,Lihu Dong*()   

  1. Key Laboratory of Sustainable Forest Ecosystem Management of Ministry of Education College of Forestry, Northeast Forestry University Harbin 150040
  • Received:2026-03-02 Revised:2026-04-20 Online:2026-11-10 Published:2026-11-20
  • Contact: Lihu Dong E-mail:donglihu2006@163.com

摘要:

目的: 量化气候与竞争及其交互作用对长白落叶松枝条基径和枝长生长的影响规律,构建兼顾生物学解释与预测能力的枝条生长模型,为基于树冠动态的抚育间伐和整枝决策提供科学依据。方法: 2023年,在吉林省辽源市林业局长白落叶松人工纯林中设置10块标准地并选伐30株解析木,获取枝条基径和枝长生长数据;对对象木周围竞争木钻取生长锥,重建竞争指数序列,并逐年匹配气候因子。基于对数线性模型,考虑样地和样木水平随机效应,构建枝条基径和枝长年生长量模型。采用刀切法检验模型预测能力,评估不同校正样本量对预测精度的影响,并利用层次分割法量化各类因子的相对贡献。结果: 1) 枝条基径和枝长生长均表现出显著的非线性年龄效应。枝条基径生长随树木年龄和枝条年龄增加均呈先增后减趋势,峰值分别出现在约32年树木年龄和4年枝条年龄;枝长生长总体随枝条年龄增加而下降,表明枝条在早期生长空间相对充足时枝长生长较快。前期基径和树高年生长量对枝条生长具有显著促进作用,而竞争增强和不利气候条件则表现为抑制效应。2) 气候与竞争存在显著交互作用,在较强气候胁迫条件下,竞争对枝条生长的抑制效应增强,高竞争水平亦会放大气候胁迫造成的生长差异,表现出明显的“双重胁迫”叠加效应;立地条件对竞争效应具有调节作用,且在枝条基径和枝长生长的调节方向上存在差异。3) 引入随机效应后,模型拟合与预测精度明显提高,其中,样木单水平混合效应模型在保持较高精度的同时结构更简洁,可作为枝条生长预测最优模型。校正抽样结果表明,随校正枝条数量增加,枝条基径和枝长预测误差总体下降且不确定性收敛,抽取4~5个枝条即可显著提高预测稳定性,低龄和高龄枝条对校正样本量更敏感。4) 层次分割结果显示,枝条基径生长主要由枝条自身因子主导,贡献率约86%,其中前期基径和枝条年龄贡献最大,说明基径增长主要受自身大小和发育阶段影响;枝长生长的驱动因子贡献更为均衡,枝条因子、单木因子和竞争因子贡献率分别约为63%、26%和6%,表明枝长生长更易受单木生长势和空间竞争的调控。结论: 气候与竞争及其交互作用是影响长白落叶松枝条生长的重要驱动因素,样木单水平混合效应模型能够较好地模拟和预测枝条生长过程。枝条基径生长模型可为判断树干不同高度潜在节子大小和确定整枝方案提供依据,枝长生长模型可用于估算树冠半径变化和树冠重叠程度,为林分密度调控、间伐时间和强度优化提供参考。

关键词: 落叶松人工林, 枝条生长, 气候变化, 竞争效应, 混合效应模型

Abstract:

Objective: This study aims to quantify the effects of climate, competition, and their interactions on annual branch base diameter and length growth of Larix olgensis, and develop branch growth models with both biological interpretability and predictive ability to support pruning and thinning decisions. Method: In 2023, 10 standard plots were established in pure L. olgensis plantations in the Liaoyuan Forestry Bureau, Jilin Province, and 30 sample trees were felled to obtain branch diameter and length growth data. Increment cores were extracted from neighboring competitor trees to reconstruct time series of competition indices, which were then matched with annual climate variables. Based on logarithmic linear models, annual branch diameter and length growth models were developed by incorporating random effects at the plot and tree levels. Model predictive ability was evaluated using a jackknife approach, the effects of different calibration sample sizes on prediction accuracy were assessed, and hierarchical partitioning was used to quantify the relative contributions of different factor groups. Result: 1) Both branch base diameter and length growth showed significant nonlinear age effects. Branch base diameter growth followed a unimodal pattern with tree age and branch age, peaking at approximately 32 years of tree age and 4 years of branch age, respectively. Branch length growth generally decreased with increasing branch age, indicating that branch elongation was faster during the early stage when growing space was relatively sufficient. Prior branch base diameter and tree height increment significantly promoted branch growth, while intensified competition and unfavorable climatic conditions had inhibitory effects. 2) There was a significant climate–competition interactions. Under stronger climatic stress, the inhibitory effect of competition on branch growth increased, and high competition also amplified growth differences caused by climatic stress, indicating a clear combined double-stress effect. Site conditions further regulated the effect of competition, and the direction of this regulation differed between branch diameter and length growth. 3) After introducing random effects, the model fitting and prediction accuracy were markedly improved. Among them, the tree-level mixed-effects model maintained high accuracy with a simpler structure and was able to be used as the optimal model for branch growth prediction. Calibration analyses showed that prediction errors for both branch base diameter and length generally decreased and uncertainty converges as the number of calibration branches increased. Sampling 4–5 branches was sufficient to substantially improve prediction stability, and young and old branches were more sensitive to calibration sample size. 4) Hierarchical partitioning showed that branch base diameter growth was mainly dominated by branch-level factors, with a contribution of approximately 86%, among them, prior branch base diameter and branch age contributed most, indicating that base diameter growth was primarily affected by branch size and developmental stage. In contrast, branch length growth showed a more balanced contribution among driving factors, with branch-, tree-, and competition factors contributing approximately 63%, 26%, and 6%, respectively, suggesting that length growth was more sensitive to tree vigor and spatial competition. Conclusion: Climate, competition, and their interactions are important drivers of branch growth, and the tree-level mixed-effect model can effectively simulate and predict branch growth dynamics. The branch base diameter growth model can provide a basis for assessing potential knot size at different stem heights and determining pruning schemes, whereas the branch length growth model can be used to estimate changes in crown radius and crown overlap, thereby supporting stand density regulation and optimization of thinning timing and intensity.

Key words: Larix olgensis plantation, branch growth, climate change, competition effect, mixed-effects model

中图分类号: