新疆甘家湖梭梭林国家级自然保护区植被群落多样性及空间分布格局

doi:10.12118/j.issn.1000-6060.2024.133

摘要/Abstract

摘要：

新疆甘家湖梭梭林国家级自然保护区位于典型的绿洲-荒漠过渡带，明确保护区内主要植被类型及分布有助于生物多样性的保护与恢复。以2022年同期遥感影像为样点布设依据，在2023年8月进行实地调查，利用双向指示种分析法（TWINSPAN）对保护区内植被样方进行分类，计算各植被类型的α多样性指数，随后构建随机森林模型以评估植被样方的聚类结果并可视化。结果表明：（1）在甘家湖保护区，草本植物以猪毛菜属（Salsola spp.）分布最为广泛，灌木以梭梭（Haloxylon ammodendron）、碱蓬属（Suaeda spp.）、红砂（Reaumuria songarica）、角果藜（Ceratocarpus arenarius）等为这一地区主要的植被组成。（2）TWINSPAN将62个植被样方划分成14个类型，其中猪毛菜属+碱蓬属群丛（Ass. Salsola spp.+Suaeda spp.，G5）所占面积最大，且植被多样性最高；猪毛菜属+角果藜群丛（Ass. Salsola spp.+Ceratocarpus arenarius，G7）物种多样性较高，同时面积占比较大；单一植被型群丛（G14）Shannon-Wiener多样性指数与Simpson优势度指数均为最低，Pielou均匀度指数为1，在保护区中主要体现为农田和盐渍化极高的裸地。（3）随机森林模型的总体精度达到87.10%，Kappa系数达到0.8553，错分误差与漏分误差相对较小。（4）梭梭和白梭梭集中分布在保护区西北部核心区以及中偏南部的缓冲区。研究结果可为保护区后续精准管理奠定基础。

关键词: 植被类型, 空间分布, 双向指示种分析法, 物种多样性, 随机森林

Abstract:

The Ganjiahu Saxoul National Nature Reserve in Xinjiang, China, is situated in a typical oasis-desert transition zone. Clearly identifying the main vegetation types and their distribution within the reserve is essential for biodiversity protection and restoration. Based on remote sensing images from 2022, a field survey was conducted in August 2023. Vegetation samples from the Ganjiahu Saxoul National Nature Reserve were classified using the two-way indicator species analysis (TWINSPAN) method. Alpha diversity indices of each vegetation type were calculated, followed by the development of a random forest model to evaluate and visualize the clustering results of vegetation quadrats. The results are as follows: (1) In the Ganjiahu Saxoul National Nature Reserve, the dominant herbaceous plants are primarily Salsola spp., while the principal shrubs include Haloxylon ammodendron, Suaeda spp., Reaumuria songarica, and Ceratocarpus arenarius. (2) Sixty-two vegetation sample plots were categorized into 14 types in TWINSPAN. Among these, the Ass. Salsola spp.+Suaeda spp. (G5) covered the largest area and exhibited the highest vegetation diversity. The Ass. Salsola spp. + Ceratocarpus arenarius (G7) showed both high species diversity and extensive coverage. The G14 type represented a single vegetation community with the lowest Shannon-Wiener and Simpson’s diversity indices and had a Pielou’s evenness index of 1, predominantly comprising farmland and highly saline bare land in the Ganjiahu Saxoul National Nature Reserve. (3) The random forest model achieved an overall accuracy of 87.10%, with a Kappa coefficient of 0.8553, and demonstrated relatively low commission and omission errors. (4) H. ammodendron and H. persicum are mainly distributed in the core area in the northwest and the buffer zone extending from the central to southern parts of the Ganjiahu Saxoul National Nature Reserve. These findings provide a foundation for future precise management of the Ganjiahu Saxoul National Nature Reserve.

Key words: vegetation types, spatial distribution, two-way indicator species analysis method, species diversity, random forest

王安琪, 张钰, 刘佳伟, 赵威, 刘馨雨, 李凯. 新疆甘家湖梭梭林国家级自然保护区植被群落多样性及空间分布格局[J]. 干旱区地理, 2024, 47(11): 1841-1851.

WANG Anqi, ZHANG Yu, LIU Jiawei, ZHAO Wei, LIU Xinyu, LI Kai. Vegetation community diversity and spatial distribution pattern in Ganjiahu Saxoul National Nature Reserve, Xinjiang[J]. Arid Land Geography, 2024, 47(11): 1841-1851.

图/表 7

图1

表1

图2

表2

图3

表3

图4

参考文献 39

[1]	方精云, 郭柯, 王国宏, 等. 《中国植被志》的植被分类系统、植被类型划分及编排体系[J]. 植物生态学报, 2020, 44(2): 96-110. doi: 10.17521/cjpe.2019.0259
	[Fang Jingyun, Guo Ke, Wang Guohong, et al. Vegetation classification system and classification of vegetation types used for the compilation of “Vegetation of China”[J]. Chinese Journal of Plant Ecology, 2020, 44(2): 96-110.] doi: 10.17521/cjpe.2019.0259
[2]	Li C J, Fu B J, Wang S, et al. Drivers and impacts of changes in China’s drylands[J]. Nature Reviews Earth & Environment, 2021, 2(12): 858-873.
[3]	刘炳强. 甘家湖荒漠林区环境恶化原因及治理对策[J]. 防护林科技, 2014(10): 85, 126.
	[Liu Bingqiang. Causes of environmental degradation and control strategies in the Ganjiahu desert forest area[J]. Protection Forest Science and Technology, 2014(10): 85, 126.]
[4]	陈雨婷, 马松梅, 张丹, 等. 新疆同域分布梭梭和白梭梭多样性格局及其形成机制[J]. 植物生态学报, 2024, 48(1): 56-67. doi: 10.17521/cjpe.2022.0343
	[Chen Yuting, Ma Songmei, Zhang Dan, et al. Diversity pattern and formation mechanism of sympatric Haloxylon ammodendron and Haloxylon persicum in Xinjiang, China[J]. Chinese Journal of Plant Ecology, 2024, 48(1): 56-67.] doi: 10.17521/cjpe.2022.0343
[5]	俞阗, 张君萍, 马洪雨, 等. 梭梭抗逆相关研究进展[J]. 新疆农业科学, 2009, 46(4): 907-912.
	[Yu Tian, Zhang Junping, Ma Hongyu, et al. Advances of study on the adversity-resistance of Haloxylon[J]. Xinjiang Agricultural Sciences, 2009, 46(4): 907-912.]
[6]	蔡新斌, 吴俊侠. 甘家湖自然保护区白梭梭种群特征与动态分析[J]. 干旱区资源与环境, 2016, 30(7): 90-94.
	[Cai Xinbin, Wu Junxia. Characteristics and dynamics analysis of Haloxylon persicum populations in the Nature Reserve of Gan Jia Lake[J]. Journal of Arid Land Resources and Environment, 2016, 30(7): 90-94.]
[7]	李引湣. 近三十年来新疆甘家湖自然保护区梭梭林退化过程研究[D]. 乌鲁木齐: 新疆师范大学, 2013.
	[Li Yinmin. Last three decades Haloxylon forest degradation process of Ganjiahu Nature Reserve in Xinjiang[D]. Urumqi: Xinjiang Normal University, 2013.]
[8]	Hill M O. TWINSPAN: A fortran program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes[D]. New York: Cornell University, 1979.
[9]	Sarker S K, Nur-Un-Nabi M, Mohasinul H M, et al. Tree assemblages and diversity patterns in tropical Juri forest, Bangladesh[J]. Journal of Forestry Research, 2015, 26(1): 159-169.
[10]	Swierkosz K, Reczyńska K, Boublík K. Variability of Abies alba-dominated forests in central Europe[J]. Open Life Sciences, 2014, 9(5): 495-518.
[11]	Zhu Y J, Shan D, Wang B Z, et al. Floristic features and vegetation classification of the Hulun Buir steppe in north China: Geography and climate-driven steppe diversification[J]. Global Ecology and Conservation, 2019, 20: e00741, doi: 10.1016/j.gecco.2019.e00741.
[12]	Nkosi S E, Barrett A S, Brown L R. Vegetation ecology of the Nooitgedacht section of Loskop Dam Nature Reserve, Mpumalanga[J]. South African Journal of Botany, 2016, 105: 79-88.
[13]	Sallam H, Alzain M N, Abuzaid A O, et al. Wild plant diversity and soil characteristics of desert roadside vegetation in the eastern desert[J]. Diversity, 2023, 15(7): 874, doi: 10.3390/d15070874.
[14]	Zeng Y, Zhao C Y, Shi F Z, et al. Impact of groundwater depth and soil salinity on riparian plant diversity and distribution in an arid area of China[J]. Scientific Reports, 2020, 10(1): 7272, doi: 10.1038/s41598-020-64045-w. pmid: 32350302
[15]	Manhas R K, Gautam M K, Kumari D. Two way indicator species analysis (TWINSPAN) of the herbaceous vegetation in an inland wetland ecosystem of Doon Valley Himalaya, India[J]. Journal of Wetlands Ecology, 2009, 3: 23-31.
[16]	Rollet R, Benie G B, Li W, et al. Image classification algorithm based on the RBF neural network and K-means[J]. International Journal of Remote Sensing, 1998, 19(15): 3003-3009.
[17]	文铜, 柳小妮, 纪童, 等. 基于植被指数的三江源高寒草地植物分类与识别方法研究[J]. 草地学报, 2022, 30(7): 1811-1818. doi: 10.11733/j.issn.1007-0435.2022.07.023
	[Wen Tong, Liu Xiaoni, Ji Tong, et al. Studying on plant classification and recognition method for Three-River Source alpine grassland plant based on vegetation index[J]. Acta Agrestia Sinica, 2022, 30(7): 1811-1818.] doi: 10.11733/j.issn.1007-0435.2022.07.023
[18]	Belgiu M, Drǎguţ L. Random forest in remote sensing: A review of applications and future directions[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 114: 24-31.
[19]	Zhou J H, Lai L M, Guan T Y, et al. Comparison modeling for alpine vegetation distribution in an arid area[J]. Environmental Monitoring and Assessment, 2016, 188(7): 408, doi: 10.1007/s10661-016-5417-x. pmid: 27307276
[20]	Han F S, Fu G, Yu C Q, et al. Modeling nutrition quality and storage of forage using climate data and normalized-difference vegetation index in alpine grasslands[J]. Remote Sensing, 2022, 14(14): 3410, doi: 10.3390/rs14143410.
[21]	Caddy-Retalic S, Wardle G M, Leitch E J, et al. Vegetation change along a Mediterranean to arid zone bioclimatic gradient reveals scale-dependent ecotone patterning[J]. Australian Journal of Botany, 2020, 68(8): 574-586.
[22]	王国宏, 方精云, 郭柯, 等. 《中国植被志》研编内容与规范[J]. 植物生态学报, 2020, 44(2): 128-178. doi: 10.17521/cjpe.2019.0272
	[Wang Guohong, Fang Jingyun, Guo Ke, et al. Contents and protocols for the classification and description of Vegetation Formations, Alliances and Associations of “Vegetation of China”[J]. Chinese Journal of Plant Ecology, 2020, 44(2): 128-178.] doi: 10.17521/cjpe.2019.0272
[23]	刘丽燕, 蔡新斌, 江晓珩, 等. 甘家湖梭梭林国家级自然保护区植被类型和特征[J]. 防护林科技, 2016(6): 54-57.
	[Liu Liyan, Cai Xinbin, Jiang Xiaoheng, et al. Types and characteristics of vegetation in Ganjiahu Haloxylon ammodendron Forest National Nature Reserve[J]. Protection Forest Science and Technology, 2016(6): 54-57.]
[24]	马倩, 楚新正, 孙慧兰, 等. 新疆甘家湖梭梭林绿洲-荒漠过渡带土地利用/覆被动态变化分析[J]. 新疆师范大学学报(自然科学版), 2014, 33(1): 8-13, 37.
	[Ma Qian, Chu Xinzheng, Sun Huilan, et al. Study on land use/cover dynamic change in Ganjia Lake Ecotone[J]. Journal of Xinjiang Normal University (Natural Sciences Edition), 2014, 33(1): 8-13, 37.]
[25]	王育松, 上官铁梁. 关于重要值计算方法的若干问题[J]. 山西大学学报(自然科学版), 2010, 33(2): 312-316.
	[Wang Yusong, Shangguan Tieliang. Discussion on calculating method of important values[J]. Journal of Shanxi University (Natural Sciences Edition), 2010, 33(2): 312-316.]
[26]	Zang S, Cao Q, Keremu A, et al. Food patch particularity and forging strategy of reintroduced Przewalski’s horse in north Xinjiang, China[J]. Turkish Journal of Zoology, 2017, 41: 924-930.
[27]	李俊晓, 李朝奎, 殷智慧. 基于ArcGIS的克里金插值方法及其应用[J]. 测绘通报, 2013(9): 87-90.
	[Li Junxiao, Li Chaokui, Yin Zhihui. ArcGIS based Kriging interpolation method and its application[J]. Bulletin of Surveying and Mapping, 2013(9): 87-90.]
[28]	刘丽燕, 蔡新斌, 江晓珩, 等. 新疆甘家湖梭梭林自然保护区植物组成及地理成分分析[J]. 植物科学学报, 2016, 34(5): 695-704.
	[Liu Liyan, Cai Xinbin, Jiang Xiaoheng, et al. Species composition and floral components of vascular plants in Ganjiahu Nature Reserve of Xinjiang[J]. Plant Science Journal, 2016, 34(5): 695-704.]
[29]	高晓宇, 郝海超, 张雪琪, 等. 中国西北干旱区植被水分利用效率变化对气象要素的响应——以新疆为例[J]. 干旱区地理, 2023, 46(7): 1111-1120. doi: 10.12118/j.issn.1000-6060.2022.545
	[Gao Xiaoyu, Hao Haichao, Zhang Xueqi, et al. Responses of vegetation water use efficiency to meteorological factors in arid areas of northwest China: A case of Xinjiang[J]. Arid Land Geography, 2023, 46(7): 1111-1120.] doi: 10.12118/j.issn.1000-6060.2022.545
[30]	Puigdefabregas J, Sole A, Gutierrez L, et al. Scales and processes of water and sediment redistribution in drylands: Results from the Rambla Honda field site in southeast Spain[J]. Earth-Science Reviews, 1999, 48(1-2): 39-70.
[31]	徐先英, 严平, 郭树江, 等. 干旱荒漠区绿洲边缘典型固沙灌木的降水截留特征[J]. 中国沙漠, 2013, 33(1): 141-145. doi: 10.7522/j.issn.1000-694X.2013.00020
	[Xu Xianying, Yan Ping, Guo Shujiang, et al. The interception loss of rainfall by three sand-fixing shrubs at the fringe of Minqin Oasis[J]. Journal of Desert Research, 2013, 33(1): 141-145.] doi: 10.7522/j.issn.1000-694X.2013.00020
[32]	Li X R, Xiao H L, Zhang J G, et al. Long-term ecosystem effects of sand-binding vegetation in the Tengger Desert, northern China[J]. Restoration Ecology, 2004, 3(12): 376-390.
[33]	李引湣, 曲宁. 甘家湖梭梭林的变化浅析[J]. 和田师范专科学校学报, 2014(5): 77-82.
	[Li Yinmin, Qu Ning. A brief analysis of the changes in Haloxylon forest in Ganjiahu[J]. Journal of Hetian Normal College, 2014(5): 77-82.]
[34]	朱家龙, 周智彬, 王立生, 等. 免灌人工梭梭林生长与土壤水分变化的耦合关系[J]. 干旱区地理, 2022, 45(5): 1579-1590. doi: 10.12118/j.issn.1000-6060.2022.054
	[Zhu Jialong, Zhou Zhibin, Wang Lisheng, et al. Coupling relationship between growth and soil moisture change of Haloxylon ammodendron plantation without irrigation[J]. Arid Land Geography, 2022, 45(5): 1579-1590.] doi: 10.12118/j.issn.1000-6060.2022.054
[35]	曹家睿. 甘家湖白梭梭空间格局分析[J]. 南方农业, 2020, 14(27): 186-188, 218.
	[Cao Jiarui. Analysis of the spatial pattern of Haloxylon persicum in Ganjiahu[J]. South China Agriculture, 2020, 14(27): 186-188, 218.]
[36]	Abdollahi A, Pradhan B, Alamri A. Regional-scale analysis of vegetation dynamics using satellite data and machine learning algorithms: A multi-factorial approach[J]. International Journal on Smart Sensing and Intelligent Systems, 2023, 16(1): 264378840, doi: 10.2478/ijssis-2023-0013.
[37]	Tikuye B G, Rusnak M, Manjunatha B R, et al. Land use and land cover change detection using the random forest approach: The case of the upper Blue Nile River Basin, Ethiopia[J]. Global Challenges, 2023, 7(10): 2300155, doi: 10.1002/gch2.202300155.
[38]	Francis R J, Kingsford R T, Moseby K, et al. Tracking landscape scale vegetation change in the arid zone by integrating ground, drone and satellite data[J]. Remote Sensing in Ecology and Conservation, 2023, 10(3): 374-387.
[39]	Barrett B, Raab C, Cawkwell F, et al. Upland vegetation mapping using random forests with optical and radar satellite data[J]. Remote Sensing in Ecology and Conservation, 2016, 2(4): 212-231. doi: 10.1002/rse2.32 pmid: 31423326

序号	植物种类	拉丁学名	平均高度/cm	平均盖度/%	平均重要值
1	猪毛菜属	Salsola spp.	12.49	9.80	31.02
2	梭梭	Haloxylon ammodendron	195.74	42.39	9.70
3	碱蓬属	Suaeda spp.	30.24	17.26	8.27
4	芦苇	Phragmites australis	45.59	8.29	6.68
5	红砂	Reaumuria songarica	49.53	19.66	4.88
6	角果藜	Ceratocarpus arenarius	6.66	7.80	4.67
7	多枝柽柳	Tamarix ramosissima	166.40	54.40	4.44
8	蒿属	Artemisia spp.	42.52	23.87	4.40
9	盐节木	Halocnemum strobilaceum	19.85	27.46	4.33
10	戈壁藜	Iljinia regelii	31.83	6.78	3.30
11	黑果枸杞	Lycium ruthenicum	23.13	19.88	3.23
12	白梭梭	Haloxylon persicum	131.08	25.96	2.99
13	骆驼刺	Alhagi camelorum	41.89	14.33	2.45
14	花花柴	Karelinia caspia	39.00	5.55	1.26
15	小果白刺	Nitraria sibirica	30.38	26.50	1.11
16	胡杨	Populus euphratica	235.00	45.00	0.93
17	驼绒藜	Krascheninnikovia ceratoides	53.11	13.89	0.75
18	铃铛刺	Caragana halodendron	133.00	33.40	0.73
19	沙拐枣属	Calligonum spp.	37.60	17.40	0.72
20	盐爪爪	Kalidium foliatum	29.60	9.00	0.66
21	盐穗木	Halostachys caspica	78.75	47.00	0.45
22	雾冰藜	Grubovia dasyphylla	3.00	1.67	0.44
23	抱茎独行菜	Lepidium perfoliatum	41.00	9.17	0.39
24	大叶补血草	Limonium gmelinii	51.83	9.67	0.38
25	芨芨草	Neotrinia splendens	78.75	18.25	0.36
26	甘草	Glycyrrhiza uralensis	25.00	20.00	0.28
27	白麻	Apocynum pictum	90.00	17.00	0.26
28	东方旱麦草	Eremopyrum orientale	8.00	10.00	0.24
29	沙蓬	Agriophyllum pungens	29.50	3.00	0.20
30	蓝刺头	Echinops sphaerocephalus	16.00	1.50	0.18
31	肉苁蓉	Cistanche deserticola	18.00	1.00	0.13
32	盐生假木贼	Anabasis salsa	13.50	4.00	0.13
33	荠	Capsella bursa-pastoris	6.00	1.00	0.04

群丛编号	样方数量	Shannon-Wiener 多样性指数	Simpson优势度	Pielou均匀度指数
G1	4	1.354	0.594	0.565
G2	4	1.643	0.745	0.685
G3	2	1.475	0.652	0.640
G4	5	1.661	0.774	0.668
G5	10	1.864	0.798	0.645
G6	3	1.437	0.727	0.739
G7	4	1.748	0.792	0.703
G8	1	1.304	0.700	0.810
G9	13	1.339	0.664	0.558
G10	2	1.409	0.716	0.677
G11	4	1.593	0.760	0.641
G12	4	1.782	0.775	0.658
G13	3	1.305	0.706	0.729
G14	3	0.693	0.500	1.000

群丛编号	面积/km²	面积占比/%	生产者精度/%	使用者精度/%
G1	28.50	4.64	75.00	100.00
G2	116.15	18.91	100.00	80.00
G3	18.45	3.00	100.00	100.00
G4	59.31	9.66	60.00	75.00
G5	124.87	20.33	90.00	69.23
G6	26.07	4.25	100.00	100.00
G7	83.71	13.63	100.00	100.00
G8	2.84	0.46	100.00	100.00
G9	81.26	13.23	84.62	91.67
G10	6.63	1.08	100.00	66.67
G11	17.05	2.78	100.00	100.00
G12	1.72	0.28	75.00	100.00
G13	32.28	5.26	100.00	100.00
G14	15.39	2.51	66.67	100.00