古尔班通古特沙漠南缘梭梭林分距离结构模拟

doi:10.13466/j.cnki.lyzygl.2020.04.008

林业资源管理 ›› 2020›› Issue (4): 50-57.doi: 10.13466/j.cnki.lyzygl.2020.04.008

古尔班通古特沙漠南缘梭梭林分距离结构模拟

汪少华¹^,²(), 黄炳智³, 牛攀新²

1.湖南文理学院,湖南常德 415000
2.石河子大学农学院,新疆石河子 832000
3.广西盛麒科技有限公司,南宁 530000

收稿日期:2020-05-15 修回日期:2020-07-15 出版日期:2020-08-28 发布日期:2020-10-10
作者简介:汪少华(1982-),男,湖南常德人,讲师,博士,主要从事森林资源调查与评估方面研究。Email: wsh82627@126.com
基金资助:
国家自然科学基金项目(31660203);国家自然科学基金项目(31660196)

The Simulation of Stand Distance Structure of Haloxylon ammodendron on the Southern Edge of Gurbantunggut Desert,Northwest China

WANG Shaohua¹^,²(), HUANG Bingzhi³, NIU Panxin²

1. Hunan University of Arts and Sciences,Changde Hunan 415000,China
2. Agricultural College of Shehezi University,Shihezi Xinjiang 832000,China
3. Guangxi Shengqi Technology Co.Ltd,Nanning 530000,China

Received:2020-05-15 Revised:2020-07-15 Online:2020-08-28 Published:2020-10-10

摘要/Abstract

摘要：

以天然梭梭林分的距离结构情况为研究对象,通过对梭梭林分最近邻距离概率分布模型的构建,建立预估梭梭林木距离的模型,为荒漠树种林分空间结构的分析与重建提供方法支持。利用Stochastic Geometry 4.1软件随机生成均匀、随机和团状3种分布类型的模拟样地,使用Matlab软件对模拟样地中的目标点以及实测样地的林分进行最近邻距离分布模型拟合,并对模型进行了拟合参数估计和拟合效果的Kolmogorov-Smirnov(K-S)检验。结果表明:模拟样地和实测样地的最近邻距离分布可以用两参数威布尔分布(Weibull-2)模型进行拟合,拟合效果较好,其中模拟样地和实测样地距离分布拟合的R²值均在0.883~0.997之间,均方根误差(RMSE)取值分别在0.003~0.052与0.007~0.041之间,D值均小于D₀_.₀₅值,结果服从原定假设。因此,Weibull-2分布可以用来描述天然梭梭林分距离结构的分布,且表现出了良好的适应性,说明Weibull-2分布可以用来对不同树种的林分空间结构进行描述。

关键词: 梭梭, Weibull分布模型, 最近邻距离, 距离结构

Abstract:

This study takes the distance structure of the natural Haloxylon ammodendron stand as the research object.Through the construction of the nearest neighbor distance probability distribution model of the Haloxylon ammodendron stand,an estimated distance model independent of the tree species will be established,which will support the analysis of the spatial structure of different tree species in the forestry investigations.In this study,we use Matlab to generate three simulated plots of uniform,random and mass distribution types,and fit the nearest neighbor distance distribution model for the target points in the simulated plots and the stands of the measured plots.Through the Kolmogorov-Smirnov(K-S) test of the model fitting parameters,the results show that the simulation sample is simulated.The nearest neighbor distribution of ground and measured plots can be fitted with the Weibull distribution(Weibull-2) model with sound fitting eflects,in which the R² value of simulated and measured plots is between 0.8832 to 0.9966 and the mean square root error(RMSE) is between 0.0034 to 0.0516 and 0.0066 to 0.0414.The value of D is less than the D_0.05 value as the result meets the original hypothesis.Therefore,the Weibull-2 distribution model can be used to describe the distribution of the distance structure of the natural Haloxylon ammodendron,which shows good adaptability.It shows that the Weibull-2 distribution model can be used to describe the spatial structure of the stand of different tree species.

Key words: Haloxylon ammodendron, Weibull distribution model, nearest neighbor distance, distance structure

中图分类号:

S791.247

汪少华, 黄炳智, 牛攀新. 古尔班通古特沙漠南缘梭梭林分距离结构模拟[J]. 林业资源管理, 2020,(4): 50-57.

WANG Shaohua, HUANG Bingzhi, NIU Panxin. The Simulation of Stand Distance Structure of Haloxylon ammodendron on the Southern Edge of Gurbantunggut Desert,Northwest China[J]. FOREST RESOURCES WANAGEMENT, 2020,(4): 50-57.

图/表 6

表1

图1

图2

表2

模拟林分最近邻距离分布模型参数值

种群分类格局类型	Nr	Weibull分布参数			R²	RMSE	$D^$	D₀_.₀₅
种群分类格局类型	Nr	a	b	c	R²	RMSE	$D^$	D₀_.₀₅
均匀分布A	1	0	4.122	1.931	0.996	0.003	0.094	0.172
	2	0	3.619	1.516	0.979	0.010	0.076	0.172
	3	0	4.649	1.972	0.996	0.003	0.087	0.172
	4	0	4.638	1.997	0.889	0.016	0.119	0.172
	5	0	4.732	1.577	0.955	0.010	0.098	0.172
	6	0	3.819	2.269	0.971	0.013	0.096	0.172
	7	0	4.308	1.707	0.955	0.014	0.140	0.172
	8	0	4.214	1.900	0.9831	0.0066	0.095	0.1723
	9	0	4.117	1.845	0.9785	0.0088	0.0868	0.1723
	10	0	4.041	1.600	0.8859	0.0161	0.0968	0.1723
随机分布B	11	0	4.017	1.637	0.9115	0.0169	0.1573	0.1723
	12	0	3.501	2.690	0.9473	0.0238	0.1458	0.1723
	13	0	4.169	1.674	0.9877	0.0072	0.0886	0.1723
	14	0	4.084	1.606	0.8832	0.0167	0.1151	0.1723
	15	0	4.069	1.788	0.9563	0.0129	0.0901	0.1723
	16	0	4.605	1.905	0.9497	0.0121	0.0955	0.1723
	17	0	3.670	2.155	0.9784	0.0126	0.1019	0.1723
	18	0	3.615	1.962	0.9769	0.013	0.1224	0.1723
	19	0	3.744	1.999	0.9924	0.0058	0.073	0.1723
	20	0	5.169	1.817	0.9759	0.0066	0.0784	0.1723
团状分布C	21	0	0.8586	1.493	0.9862	0.0312	0.1024	0.1182
	22	0	0.9407	1.362	0.9818	0.0307	0.0826	0.1116
	23	0	1.0410	1.615	0.9833	0.0293	0.1032	0.1191
	24	0	0.9878	1.491	0.9881	0.0273	0.0732	0.1097
	25	0	1.0600	1.611	0.9800	0.0338	0.0996	0.1191
	26	0	0.9360	1.832	0.9707	0.0427	0.1076	0.1127
	27	0	1.0400	1.669	0.9801	0.0336	0.0875	0.1205
	28	0	0.8683	1.536	0.9666	0.0516	0.117	0.142
	29	0	0.8491	1.634	0.9754	0.0487	0.0614	0.1034
	30	0	0.9277	1.470	0.9799	0.0396	0.0798	0.1382

表2

表3

实测林分最近邻距离分布模型参数值

Nr	Weibull分布参数			R²	RMSE	$D^$	D₀_.₀₅
Nr	a	b	c	R²	RMSE	$D^$	D₀_.₀₅
1	0	2.92	1.901	0.9808	0.0141	0.0948	0.1469
2	0	2.374	1.654	0.9904	0.011	0.0773	0.0902
3	0	3.67	1.838	0.9279	0.0216	0.179	0.1984
4	0	3.995	1.692	0.9579	0.0142	0.0749	0.1028
5	0	3.126	1.957	0.9956	0.0066	0.1185	0.1187
6	0	4.002	2.498	0.9888	0.0414	0.1883	0.1942
7	0	1.638	1.212	0.9298	0.0374	0.1025	0.104
8	0	2.109	2.064	0.9931	0.0124	0.0366	0.0666
9	0	2.322	1.899	0.9966	0.0072	0.032	0.0693
10	0	2.436	2.098	0.9945	0.0101	0.0297	0.0804

表3

图3

参考文献 15

[1]	汪少华, 杨婷. 古尔班通古特沙漠南缘梭梭基径结构分布模拟[J]. 干旱区资源与环境, 2018,32(7):104-111.
[2]	王蒙, 李凤日. 基于抚育间伐效应的落叶松人工林直径分布动态模拟[J]. 应用生态学报, 2016,27(8):2429-2437.
[3]	张鹏, 王新杰, 许昊. 杉木幼龄林直径分布[J]. 东北林业大学学报, 2014,42(11):7-10.
[4]	Marc Palahí, Timo Pukkala, Elena Blasco, et al. Comparison of beta,Johnson's SB,Weibull and truncated Weibull functions for modeling the diameter distribution of forest stands in Catalonia(north-east of Spain)[J]. European Journal of Forest Research, 2007,126(4).
[5]	Rafał Podlaski, Michał Zasada. Comparison of selected statistical distributions for modelling the diameter distributions in near-natural Abies-Fagus forests in the Świętokrzyski National Park(Poland) [J]. European Journal of Forest Research, 2008,127(6):455-463.
[6]	Ralph L.Amateis, Colleen A.Carlson. Modeling diameter class removals for thinned loblolly pine(Pinus taeda) plantations[J]. Forest Ecology and Management, 2014,327:26-30.
[7]	李炳铁. 林分直径分布转移概率的最新理论—径阶枯损概率的估测方法[J].林业资源管理, 1992(4):72-82.
[8]	龙时胜, 曾思齐, 甘世书, 等. 基于林木多期直径测定数据的异龄林年龄估计方法Ⅱ[J]. 中南林业科技大学学报, 2019,39(6):23-29.
[9]	李金亮. 秃杉林分密度与胸径、树高、年龄之间的回归分析[J]. 林业调查规划, 2018,43(2):14-18.
[10]	方精云, 营诚. 利用Weibull分布函数预测林木的直径分布[J].北京林业大学学报, 1987(3):261-269.
[11]	孟宪宇. 使用Weibull函数对树高分布和直径分布的研究[J].北京林业大学学报, 1988(1):40-48.
[12]	黄培祐, 李启剑, 袁勤芬. 准噶尔盆地南缘梭梭群落对气候变化的响应[J]. 生态学报, 2008,28(12):6051-6059.
[13]	龚直文, 亢新刚, 杨华, 等. 长白山杨桦次生林直径结构研究[J]. 西北林学院学报, 2009,24(3):1-6.
[14]	方子兴. 韦布尔分布及其参数估计[J].林业科学研究, 1993(4):426-430.
[15]	曾德慧, 姜凤岐. 樟子松沙地人工林直径分布模拟[J]. 应用生态学报, 1997,8(3):231-234.

样地	样株数 /株	密度/ (株/hm²)	平均值 (D/H)	标准差 (D/H)	峰度 (D/H)	偏度 (D/H)
1	353	679	3.97/1.22	4.25/0.85	6.15/2.21	1.67/0.50
2	593	1140	4.01/1.42	3.34/0.71	6.70/2.57	1.76/0.23
3	253	487	5.49/1.53	2.66/0.51	3.66/2.80	0.65/-0.13
4	421	810	5.26/1.67	4.08/0.79	6.52/3.50	1.73/0.69
5	393	756	8.54/2.25	4.95/0.71	2.99/3.50	0.67/-0.46
6	379	729	6.04/1.79	4.28/0.77	7.17/2.40	1.28/-0.14
7	281	540	6.73/1.88	4.47/0.68	3.01/2.65	0.80/-0.16
8	335	644	2.75/1.03	2.75/0.51	10.74/2.59	2.49/0.57
9	558	1073	5.43/1.70	3.37/0.63	7.06/3.67	1.70/0.49
10	223	429	5.62/1.87	5.06/0.77	4.52/3.72	1.56/0.71

古尔班通古特沙漠南缘梭梭林分距离结构模拟

The Simulation of Stand Distance Structure of Haloxylon ammodendron on the Southern Edge of Gurbantunggut Desert,Northwest China

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 15

相关文章 1

编辑推荐

Metrics

本文评价