欢迎访问林草资源研究

林业资源管理 ›› 2020›› Issue (5): 89-99.doi: 10.13466/j.cnki.lyzygl.2020.05.014

• 科学研究 • 上一篇    下一篇

基于哑变量的秦巴山区天然栎类林胸径和树高生长模型研究

闵志强1, 胡云云1(), 王得军1, 孙景梅1, 李宏韬1, 李卫忠2   

  1. 1.国家林业和草原局西北调查规划设计院 旱区生态水文与灾害防治国家林业和草原局重点实验室,西安 710048
    2.西北农林科技大学 林学院,陕西 杨凌 712100
  • 收稿日期:2020-07-29 修回日期:2020-09-27 出版日期:2020-10-28 发布日期:2020-11-30
  • 通讯作者: 胡云云
  • 作者简介:闵志强(1985-),男,河南鹤壁人,高工,主要从事森林资源监测、森林经营等方面的生产实践与理论
  • 基金资助:
    陕西省林业科学研究重大项目“森林质量精准提升关键技术研究”(SXLY-2018-02)

Study on Growth Model of DBH and Height Based on Dummy Variable for Natural oak Forest in Qinba Mountain Area

MIN Zhiqiang1, HU Yunyun1(), WANG Dejun1, SUN Jingmei1, LI Hongtao1, LI Weizhong2   

  1. 1. Key Laboratory National Forestry and Grassland Administration on Ecological Hydrology and Disaster Prevention in Arid Regions,Northwest Surveying,Planning and Designing Institute of National Forestry and Grassland Administration,Xi'an 710048,China
    2. College of Forestry,Northwest A&F University,Yangling,Shaanxi 712100,China;
  • Received:2020-07-29 Revised:2020-09-27 Online:2020-10-28 Published:2020-11-30
  • Contact: HU Yunyun

摘要:

利用全国森林资源连续清查数据,选取秦巴山区天然栎类林作为研究对象,基于竞争指数确定优势木及其竞争等级;根据优势木和立地主导因子确定立地质量等级;将竞争等级和立地质量等级作为哑变量逐步参与胸径、树高生长过程拟合,确定最优生长模型。研究发现:栎类天然林的立地主导因子为海拔、坡向和土壤厚度;胸径、树高生长模型的最优基础模型均为Gompertz模型;胸径生长模型以竞争等级哑变量位于参数a,c时拟合结果最好,其中R2值和Pa值分别为0.834和98.17,较基础模型分别提高13.16%,0.59%;树高生长模型以立地质量等级哑变量位于参数a,b时拟合结果最好,其中R2值和Pa值分别为0.717和98.26,较基础模型分别提高16.59%和0.65%;当胸径和树高模型引入第二类哑变量时,模型精度虽有提升但效果不明显。分析秦巴山区不同生长条件下天然栎类林的胸径和树高生长规律,建立林分生长模型,以期为天然栎类次生林修复与林分质量提升提供重要参考。

关键词: 天然栎类林, 哑变量, 立地质量, 竞争指数, 生长模型

Abstract:

Based on the national forest resources inventory data,this study takes the natural oak forest in Qinba mountain area as the research object.The dominant trees and their competitive grades were determined based on the competition index and the site quality grade was determined according to the dominant trees and site dominant factors.The competition grade and site quality grade were used as dummy variables in the fitting of the growth process of the DBH and tree height to obtain the optimal growth model.The results showed that the dominant site factors of oak natural forest were altitude,aspect and soil thickness;Gompertz model was the optimal basic model for the DBH and tree height growth model;the best fitting results were obtained when the dummy variable of competition grade was in parameters a and c,in which R 2 value and Pa value were 0.834 and 98.17,respectively,13.16% and 0.59% higher than those of the basic model. The study found that the height growth model used Gompertz model when the dummy variables of site quality grade were in parameters a and b,the fitting results were the best.R 2 value and Pa value were 0.717 and 98.26 respectively,which were 16.59% and 0.65% higher than those of the basic model.When the second kind of dummy variable is introduced into the DBH and tree height model,the accuracy of the model is improved,but the effect is not obvious.Based on the analysis of the DBH and tree height growth rules of natural oak forest under different growth conditions in Qinba mountain area,the stand growth model was established,which can provide important reference for the restoration of natural oak secondary forest and the improvement of stand quality.

Key words: natural oak, dummy variable, site quality, competitive index, growth model

中图分类号: