不同郁闭度天山云杉林分形维数特征  

Study on fractal dimension characteristics of Picea schrenkiana var.tianschanica forests with different canopy densities

作　　者：马琪瑶 王振锡[1,2] 曲延斌 葛瑶 吕金城 郝康迪 董巍 MA Qiyao;WANG Zhenxi;QU Yanbin;GE Yao;LYU Jincheng;HAO Kangdi;DONG Wei(College of Forestry and Landscape Architecture,Xinjiang Agricultural University,Urumqi 830052,Xinjiang,China;Key Laboratory of Forestry Ecology and Industrial Technology in Arid Area of Education Department of Xinjiang Uxgur Autonomous Region,Urumqi 830052,Xinjiang,China)

机构地区：[1]新疆农业大学林学与风景园林学院,新疆乌鲁木齐830052 [2]新疆维吾尔自治区教育厅干旱区林业生态与产业技术重点实验室,新疆乌鲁木齐830052

出　　处：《中南林业科技大学学报》2022年第5期65-73,共9页Journal of Central South University of Forestry & Technology

基　　金：中央财政林业改革发展资金项目“天保工程精准监测技术验证与评价”(〔2020〕942)。

摘　　要：【目地】以天山云杉林为研究对象,选取疏、中、密3种不同郁闭度的天山云杉林典型样方,分别计算其分形维数,探究不同郁闭度下天山云杉林分形特征的稳定性,以期为新疆山区天然林森林分类提供新的思路。【方法】首先使用ArcGIS软件对无人机影像进行图形深度转换,然后在MATLAB系统下分别采用像素点覆盖法和差分盒子维法计算60个不同郁闭度天山云杉林典型样方的分形维数,最后对计算结果进行分析讨论。【结果】(1)在不同郁闭度(疏,0.20~0.39;中,0.40~0.59;密,0.60以上)下,采用像素点覆盖法,得到疏、中和密3种郁闭度下天山云杉林的平均分形维数分别为1.6230、1.6313和1.6445,说明天山云杉林的分形维数受林分郁闭度的影响不大,具有很强的稳定性。(2)在不同郁闭度下,采用差分盒子维法,得到疏、中和密3种郁闭度下天山云杉林的平均分形维数分别为2.3611、2.3573和2.3774,方差分析表明三者之间没有显著性差异,分形维数较为稳定。(3)通过对两种方法计算结果离散程度的比较,差分盒子维法离散程度较小,所得数据稳定性较高。【结论】基于两种方法计算求得的天山云杉林分形维数,能较好地反映在不同郁闭度下天山云杉林的分形特征较为稳定,且差分盒子维法优于像素点覆盖法,这为今后天山云杉林的调查研究工作提供了新的思路。【Objective】Taking Picea schrenkiana var.tianschanica forests as the research object,three typical quadrats of Picea schrenkiana var.tianschanica forests with different canopy densities were selected,and their fractal dimensions were calculated respectively to explore the stability of fractal characteristics of Picea schrenkiana var.tianschanica forests under different canopy densities,so as to provide new ideas for forest classification of natural forests in mountainous areas of Xinjiang.【Method】In this study,ArcGIS was firstly used to conduct the in-depth processing of UAV images.Secondly,the pixel coverage method and difference box dimension method were used to calculate the fractal dimension,and MATLAB software was used to calculate the fractal dimension of 60 typical spruce forest plots under different closures.Finally,the calculation results were analyzed and discussed.【Result】(1)Under different canopy densities(sparse:0.20-0.39,medium:0.40-0.59,dense:more than 0.60),the average fractal dimensions of Picea schrenkiana var.tianschanica forests under sparse,medium and dense forest canopy densities were 1.6230,1.6313 and 1.6445,respectively,according to the pixel coverage method analysis.That was,the fractal dimension of Picea schrenkiana var.tianschanica forests was stable,and it was not significantly changed with the variation of closures,which further proved that Picea schrenkiana var.tianschanica forests had good stability under different closures.(2)Under different canopy densities,the difference box dimension method analysis indicated that the average fractal dimensions of Picea schrenkiana var.tianschanica forests under sparse,medium and dense canopy densities were 2.3611,2.3573 and 2.3774,respectively.The analysis of variance showed that there was no significant difference among the three,and the fractal dimension was relatively stable.(3)By comparing the dispersion degree of the two methods,the difference box dimension method was less discrete,that was,using the difference box dimension met

关 键 词：郁闭度 天山云杉林 分形维数 盒子维 

分 类 号：S771.8[农业科学—森林工程]