文献类型：期刊文献

中文题名：大兴安岭地区兴安落叶松天然林转移矩阵模型研究

英文题名：Transition Growth Matrix Models for Mixed Larix gmelinii Natural Forestsin Daxing an Mountain

作者：王建军[1] 冉啟香[2] 张孝仙[1]

第一作者：王建军

机构：[1]中国林业科学研究院林业科技信息研究所,北京100091;[2]重庆市巫溪县林业局,重庆405800

年份：2024

卷号：39

期号：3

起止页码：1-9

中文期刊名：西北林学院学报

外文期刊名：Journal of Northwest Forestry University

收录：CSTPCD;;北大核心:【北大核心2023】；CSCD:【CSCD2023_2024】；

基金：国家自然科学基金面上项目(31570633)。

语种：中文

中文关键词：兴安落叶松天然林;转移矩阵生长模型;向上生长;进界;枯损

外文关键词：Larix gmelinii natural forest;transition matrix model;growth;recruitment;mortality

分类号：S757.2

摘要：建立内蒙古大兴安岭地区兴安落叶松天然林的径阶转移矩阵生长模型,用于预估主要树种的生长情况及林分结构参数,为森林结构与功能恢复提供参考。利用2期一类清查数据的667块兴安落叶松天然林样地数据,构建兴安落叶松、桦木、栎类、软阔的可变参数转移矩阵模型(VM)和固定参数转移矩阵模型(FM),应用构建的生长、进界、枯损3个子模型,预测短期内的断面积变化以及长期内的林分参数变化。结果表明,VM拟合结果显示,林木大小、林木多样性、林分属性、竞争指数和立地条件对林木向上生长、进界和枯损具有显著影响(P<0.05);模型的R 2均在0.82以上,MAE在18%以内,VM和FM的预测精度相差不大。FM和VM对胸高断面积的短期预测值与实际值基本无差异;FM对林木大小多样性指数、树种多样性指数、胸高断面积和林分密度的长期预测值偏高,而VM的预测结果符合理论值。研究显示,转移矩阵生长模型具有较好的解释性和预测性,采用考虑林分状态的可变参数转移矩阵模型法可以合理预测大兴安岭地区兴安落叶松的林分结构。
The diameter transfer matrix growth model of Larix gmelinii natural forest occurring in the Daxing an Mountain in Inner Mongolia was established to predict the forest growth and stand attributes,and to provide some references for restoration of forest structure and function.Based on the data of 667 sample plots of L.gmelinii natural forest from 8 th and 9 th Chinese National Continuous Forest Inventory in Inner Mongolia,we developed the variable parameter transition matrix models(VMs)and fixed parameter transition matrix models(FMs)for major species groups,namely Pine,Birch,Oak and OS.Then,the growth models,recruitment models,and mortality models were used to predict stand basal area in the next five years and stand parameters in nearly one hundred years.VM fitting indicated that the tree size,diversity,stand attributes,competition index and site conditions had significant effects on the tree growth,recruitment,and mortality(P<0.05).The R 2 of the models were above 0.82,and the values of MAE were less than 18%,but there was little difference between FMs and VMs in prediction accuracy.The results of the short-term prediction of FM for tree size diversity index,tree species diversity index,stand basal area and stand density were high,and the long-term prediction of VM was in line with the theoretical value.Therefore,the transition matrix models of L.gmelinii natural forest have good explanatory and predictive ability,in which the VM that takes into account the forest stand status provides a reasonable forecast of stand structure.