基于地理探测器的乌鲁木齐市城区扩展及影响因素分析

doi:10.12118/j.issn.1000–6060.2021.06.21

摘要/Abstract

摘要：

以西北干旱区城市新疆乌鲁木齐市为例,利用2000、2010年和2020年3期Landsat TM及OLI遥感影像数据,采用随机森林分类方法,基于250 m×250 m网格单元,计算每个网格单元中各类用地面积比率,对2000—2020年乌鲁木齐市土地覆被/土地利用（LUCC）空间特征进行分析,运用地理探测器模型进一步揭示其驱动因素,定量分析评价了2000—2020年乌鲁木齐市土地覆被/土地利用时空变化特征及响应因素。结果表明：（1）近20 a乌鲁木齐市城市化发展迅速,建设用地面积大幅增加,在2000年建成区基础上向北、东北、西北以及东南方向不断蔓延发展,主要以向外扩张式发展,也存在向城市内部填充式集约增长的趋势。（2）近20 a绿地虽出现一定程度的波动,但总体表现为增长趋势;裸地始终在朝向持续减少的趋势发展。（3）影响乌鲁木齐市土地覆被/土地利用时空变化分异的主导影响因子是归一化植被指数（NDVI）,其贡献率最高;在交互因子探测结果中,NDVI与高程共同作用时,呈现双因子增强,解释力最大。监测近20 a乌鲁木齐市土地利用的动态变化,可更好地为当前乌鲁木齐城市发展提供借鉴,同时还可对城市未来规划方向进行借鉴。

关键词: 土地利用, 随机森林算法, 网格单元, 地理探测器, 乌鲁木齐市

Abstract:

In this paper, taking Urumqi City in northwest China as an example, the spatial land cover/land use (LUCC) characteristics of Urumqi from 2000 to 2020 were analyzed on the basis of the thematic mapper (TM) and oil remote sensing image data from three phases of Landsat in 2000, 2010, and 2020. This was achieved using the random forest classification method based on a 250 m×250 m grid cell. The underlying factors were further revealed using a geographic detector model, and a quantitative analysis and an evaluation of Urumqi City from 2000 to 2020 were carried out to determine the spatial and temporal changes and response factors of LUCC over the past 20 years. The results demonstrated the following: (1) Over the past 20 years, Urumqi has been rapidly urbanized, and the land area dedicated to construction has greatly increased. On the basis of the built-up area in 2000, Urumqi has been developing to the north, northeast, northwest, and southeast, mainly in the outward expansion mode, although urban interior filling-type growth also exhibits a trend of intensive growth. (2) Over the past 20 years, the green space in Urumqi City has fluctuated to a certain extent, although there has been an increasing trend overall. At the same time, the amount of bare land has shown a trend of continuous decrease. (3) The main influencing the temporal and spatial variation of LUCC in Urumqi City were identified using a geographical detector, and the results revealed that the dominant factor with the highest contribution rate was normalized difference vegetation index (NDVI). According to the interactive factor detection results, when NDVI and digital elevation model were used together, the two enhanced each other, and the explanatory power was the greatest.

Key words: land use, random forest algorithm, grid cell, geographical detector, Urumqi City

赵永玉,阿里木江·卡斯木,高鹏文,梁洪武. 基于地理探测器的乌鲁木齐市城区扩展及影响因素分析[J]. 干旱区地理, 2021, 44(6): 1729-1739.

ZHAO Yongyu,Alimujiang KASIM,GAO Pengwen,LIANG Hongwu. Quantitative analysis of urban expansion and response factors in Urumqi City based on random forest algorithm and geographical detector[J]. Arid Land Geography, 2021, 44(6): 1729-1739.

图/表 12

表1

表2

图1

表3

图2

表4

图3

图4

表5

图5

表6

图6

参考文献 37

[1]	刘亚茹, 王聪, 严力蛟. 华北平原农区土地利用变化对生态系统服务的影响——以河南省商丘市为例[J]. 应用生态学报, 2018, 29(5):1597-1606.
	[ Liu Yaru, Wang Cong, Yan Lijiao. Impacts of land use change on ecosystem services in the agricultural area of North China Plain: A case study of Shangqiu City, Henan Province, China[J]. Chinese Journal of Applied Ecology, 2018, 29(5):1597-1606. ]
[2]	刘金勇, 孔繁花, 尹海伟, 等. 济南市土地利用变化及其对生态系统服务价值的影响[J]. 应用生态学报, 2013, 24(5):1231-1236.
	[ Liu Jinyong, Kong Fanhua, Yin Haiwei, et al. Land use change and its effects on ecosystem service value in Jinan City of Shandong Province, east China[J]. Chinese Journal of Applied Ecology, 2013, 24(5):1231-1236. ]
[3]	Gao J, Li F, Gao H, et al. The impact of land-use change on water-related ecosystem services: A study of the Guishui River Basin, Beijing, China[J]. Journal of Cleaner Production, 2016, 163(Suppl. 1):148-155.
[4]	Shen G, Yang X C, Jin Y X, et al. Land use changes in the Zoige Plateau based on the object-oriented method and their effects on landscape patterns[J]. Remote Sensing, 2019, 12(1):14, doi: 10.3390/rs12010014. doi: 10.3390/rs12010014
[5]	Genbatu Ge, Shi Z J, Zhu Y J, et al. Land use/cover classification in an arid desert-oasis mosaic landscape of China using remote sensed imagery: Performance assessment of four machine learning algorithms[J]. Global Ecology and Conservation, 2020, 22, doi: 10.1016/j.gecco.2020.e00971. doi: 10.1016/j.gecco.2020.e00971
[6]	Meghan Blumstein, Jonathan R Thompson. Land-use impacts on the quantity and configuration of ecosystem service provisioning in Massachusetts, USA[J]. Journal of Applied Ecology, 2015, 52(4):1009-1019. doi: 10.1111/1365-2664.12444
[7]	韩海青, 王旭红, 牛林芝, 等. 1992—2015年中亚五国LUCC特征及耕地驱动力研究[J]. 中国生态农业学报, 2021, 29(2):325-339.
	[ Han Haiqing, Wang Xuhong, Niu Linzhi, et al. The LUCC characteristics and driving forces of cultivated land in the five Central Asian countries from 1992 to 2015[J]. Chinese Journal of Eco-Agriculture, 2021, 29(2):325-339. ]
[8]	Zhang M Y, Zhang L, Ren X L, et al. Effect of land use and land cover change on the changes in net primary productivity in karst areas of southwest China: A case study of Huanjiang Maonan Autonomous County[J]. Journal of Resources and Ecology, 2020, 11(6):606-616.
[9]	朱增云, 阿里木江·卡斯木. 基于CA-Markov模型的呼图壁县土地利用景观格局预测研究[J]. 生态科学, 2020, 39(1):136-145.
	[ Zhu Zengyun, Kasimu Alimujiang. Prediction of land use landscape pattern in Hutubi County based on CA-Markov model[J]. Ecological Science, 2020, 39(1):136-145. ]
[10]	陈磊士, 赵俊三, 李易, 等. 基于机器学习的多源遥感影像融合土地利用分类研究[J]. 西南师范大学学报(自然科学版), 2018, 43(10):103-111.
	[ Chen Leishi, Zhao Junsan, Li Yi, et al. On land use classification by means of machine learning based on muiti-source remote sensing fusion[J]. Journal of Southwest China Normal University (Natural Science Edition), 2018, 43(10):103-111. ]
[11]	李平, 吴曼乔, 曾联明. 支持向量机技术在土地利用监测的应用研究[J]. 测绘通报, 2010(8):28-30, 67.
	[ Li Ping, Wu Manqiao, Zeng Lianming. On application of suport vector machine technology to land use monitoring[J]. Bulletin of Surveying and Mapping, 2010(8):28-30, 67. ]
[12]	李国庆, 黄菁华, 刘冠, 等. 基于Landsat8卫星影像土地利用景观破碎化研究——以陕西省延安麻塔流域为例[J]. 国土资源遥感, 2020, 32(3):121-128.
	[ Li Guoqing, Huang Jinghua, Liu Guan, et al. A study of the landscape fragmentations of land cover structure based on Landsat8 remote sensing image: A case study of Mata Watershed in Yan’an, Shaanxi Province[J]. Remote Sensing for Land and Resources, 2020, 32(3):121-128. ]
[13]	赵海莉, 王启雯, 朱立祥, 等. 基于地理探测器的欠发达地区健康资源时空分异及影响因素研究[J]. 干旱区地理, 2021, 44(2):594-603.
	[ Zhao Haili, Wang Qiwen, Zhu Lixiang, et al. Analysis of spatial-temporal evolution and influencing factors of health resources in underdeveloped areas based on geodetectors[J]. Arid Land Geography, 2021, 44(2):594-603.
[14]	郑续, 魏乐民, 郭建军, 等. 基于地理探测器的干旱区内陆河流域产水量驱动力分析——以疏勒河流域为例[J]. 干旱区地理, 2020, 43(6):1477-1485.
	[ Zheng Xu, Wei Lemin, Guo Jianjun, et al. Driving force analysis of water yield in inland river basins of arid areas based on geo-detectors: A case of the Shule River[J]. Arid Land Geography, 2020, 43(6):1477-1485. ]
[15]	吕晨, 蓝修婷, 孙威. 地理探测器方法下北京市人口空间格局变化与自然因素的关系研究[J]. 自然资源学报, 2017, 32(8):1385-1397.
	[ Lü Chen, Lan Xiuting, Sun Wei. A study on the relationship between natural factors and population distribution in Beijing using geographical detector[J]. Journal of Natural Resources, 2017, 32(8):1385-1397. ]
[16]	王劲峰, 徐成东. 地理探测器: 原理与展望[J]. 地理学报, 2017, 72(1):116-134. doi: 10.11821/dlxb201701010
	[ Wang Jinfeng, Xu Chengdong. Geodetector: Principle and prospective[J]. Acta Geographica Sinica, 2017, 72(1):116-134. ] doi: 10.11821/dlxb201701010
[17]	王冬辰, 杜培军, 苏红军, 等. 近20年大同市土地利用/覆盖遥感变化分析[J]. 干旱区资源与环境, 2015, 29(7):68-75.
	[ Wang Dongchen, Du Peijun, Su Hongjun, et al. Land use/land cover changes in Datong based on remote sensing data[J]. Journal of Arid Land Resources and Environment, 2015, 29(7):68-75. ]
[18]	Schirpke U, Kohler M, Leitinger G, et al. Future impacts of changing land-use and climate on ecosystem services of mountain grassland and their resilience[J]. Ecosystem Services, 2017, 26:79-94. doi: 10.1016/j.ecoser.2017.06.008
[19]	鄂子骥, 阿里木江·卡斯木, 买买提江·买提尼亚孜, 等. 基于网格单元的西北干旱区城市土地覆被/土地利用时空变化研究——以新疆喀什市为例[J]. 干旱区地理, 2018, 41(3):625-633.
	[ E Ziji, Kasimu Alimiujiang, Maitiniyazi Maimaitijiang, et al. Temporal and spatial variations of urban land cover/land use based on grid element in northwest arid area of China: A case of Kashgar City in Xinjiang[J]. Arid Land Geography, 2018, 41(3):625-633. ]
[20]	阿里木江•卡斯木, 玉苏普江•艾麦提. 基于ALOS数据的乌鲁木齐市绿地景观格局研究[J]. 地域研究与开发, 2012, 31(2):86-89.
	[ Kasimu Alimujiang, Aimaitia Yusupujiang. Study on spatial structure of urban greenbelt landscapes in Urumqi City based on ALOS satellite image[J]. Areal Research and Development, 2012, 31(2):86-89. ]
[21]	哈孜亚·包浪提将, 毋兆鹏, 陈学刚, 等. 乌鲁木齐市景观格局变化及驱动力分析[J]. 生态科学, 2018, 37(1):62-70.
	[ Baolangtijiang Haziya, Wu Zhaopeng, Chen Xuegang, et al. Analysis of landscape pattern change and driving force in Urumqi City[J]. Ecological Science, 2018, 37(1):62-70. ]
[22]	赵领娣, 李莎莎, 赵志博, 等. 干旱半干旱区城市生态效率时空演变及区域差异分析[J]. 干旱区地理, 2020, 43(2):449-457.
	[ Zhao Lingdi, Li Shasha, Zhao Zhibo, et al. Temporal and spatial evolution and regional difference analysis of urban ecological efficiency in arid and semiarid areas[J]. Arid Land Geography, 2020, 43(2):449-457. ]
[23]	吴志杰, 赵书河. 基于TM图像的“增强的指数型建筑用地指数”研究[J]. 国土资源遥感, 2012, 93(2):50-55.
	[ Wu Zhijie, Zhao Shuhe. A study of enhanced index-based built-up index based on Landsat TM imagery[J]. Remote Sensing for Land & Resource, 2012, 93(2):50-55. ]
[24]	Ghulam Abduwasit, Ghulam Oghlan, Maitiniyazi Maimaitijiang, et al. Remote sensing based spatial statistics to document tropical rainforest transition pathways[J]. Remote Sensing, 2015, 7(5):6257-6279. doi: 10.3390/rs70506257
[25]	王珊珊, 陈曦, 段含明, 等. 城市地表温度对土地利用/覆被变化响应的遥感研究——以乌鲁木齐为例[J]. 中国沙漠, 2012, 32(3):878-884.
	[ Wang Shanshan, Chen Xi, Duan Hanming, et al. Study on responsse of land surface temperature to land use and land cover change using remote sensing data: A case on Urumqi, China[J]. Journal of Desert Research, 2012, 32(3):878-884. ]
[26]	Sexton J O, Urban D L, Donohue M J, et al. Long-term land cover dynamics by multi-temporal classification across the Landsat-5 record[J]. Remote Sensing of Environment, 2013, 128:246-258. doi: 10.1016/j.rse.2012.10.010
[27]	赵亚杰, 王立辉, 孔祥兵, 等. 基于Sentinel-2和Landsat 8 OLI数据融合的土地利用分类研究[J]. 福建农林大学学报(自然科学版), 2020, 49(2):248-255.
	[ Zhao Yajie, Wang Lihui, Kong Xiangbing, et al. Land use classification based on data fusion of Sentinel-2 and Landsat 8 OLI[J]. Journal of Fujian Agriculture and Forestry University (Natural Science Edition), 2020, 49(2):248-255. ]
[28]	Breiman L. Random forests[J]. Machine Learning, 2001, 45(1):5-32. doi: 10.1023/A:1010933404324
[29]	董师师, 黄哲学. 随机森林理论浅析[J]. 集成技术, 2013, 2(1):1-7.
	[ Dong Shishi, Huang Zhexue. A brief theoretical overview of random forests[J]. Journal of Integration Technology, 2013, 2(1):1-7. ]
[30]	张卫春, 刘洪斌, 武伟. 基于随机森林和Sentinel-2影像数据的低山丘陵区土地利用分类——以重庆市江津区李市镇为例[J]. 长江流域资源与环境, 2019, 28(6):1334-1343.
	[ Zhang Weichun, Liu Hongbin, Wu Wei. Classification of land use in low mountain and hilly area based on random forest and Sentinel-2 satellite data: A case study of Lishi Town, Jiangjin, Chongqing[J]. Resources and Environment in the Yangtze Basin, 2019, 28(6):1334-1343. ]
[31]	谷晓天. 基于机器学习的湟水流域土地利用/土地覆被分类研究[D]. 西宁: 青海师范大学, 2018.
	[ Gu Xiaotian. Research on land use/land cover classification based on machine learning in the Huangshui River Basin[D]. Xining: Qinghai Normal University, 2018. ]
[32]	王娜, 李强子, 杜鑫, 等. 单变量特征选择的苏北地区主要农作物遥感识别[J]. 遥感学报, 2017, 21(4):519-530.
	[ Wang Na, Li Qiangzi, Du Xin, et al. Identification of main crops based on the univariate feature selection in Subei[J]. Journal of Remote Sensing, 2017, 21(4):519-530. ]
[33]	朱永森, 曾永年, 张猛. 基于HJ卫星数据与面向对象分类的土地利用/覆盖信息提取[J]. 农业工程学报, 2017, 33(14):258-265.
	[ Zhu Yongsen, Zeng Yongnian, Zhang Meng. Extract of land use/cover information based on HJ satellites data and object-oriented classification[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(14):258-265. ]
[34]	买买提江·买提尼亚孜, 阿里木江·卡斯木. 基于网格单元的乌鲁木齐市土地覆被/利用时空变化[J]. 农业工程学报, 2018, 34(1):210-216.
	[ Maitiniyazi Maimaitijiang, Kasimu Alimujiang. Spatial-temporal change of Urumqi urban land use and land cover based on grid cell approach[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(1):210-216. ]
[35]	方创琳. 天山北坡城市群可持续发展战略思路与空间布局[J]. 干旱区地理, 2019, 42(1):1-11.
	[ Fang Chuanglin. Strategic thinking and spatial layout for the sustainable development of urban agglomeration in northern slope of Tianshan Mountains[J]. Arid Land Geography, 2019, 42(1):1-11. ]
[36]	杨雅楠, 阿里木江·卡斯木. “一带一路”背景下新疆城镇交通优势度与区域经济发展水平的关系分析[J]. 干旱区地理, 2017, 40(3):680-691.
	[ Yang Ya’nan, Kasimu Alimujiang. Relationship between regional transportation advantage and development of reginal economy in Xinjiang under the background of the national strategy of the Belt and Road[J]. Arid Land Geography, 2017, 40(3):680-691. ]
[37]	王雪微, 王士君, 宋飏, 等. 交通要素驱动下的长春市土地利用时空变化[J]. 经济地理, 2015, 35(4):155-161.
	[ Wang Xuewei, Wang Shijun, Song Yang, et al. Changchun land use spatio-temporal variation under the trasportation elements’ driving[J]. Economic Geography, 2015, 35(4):155-161. ]

年份	影像选取时间（年-月-日）	条带号（列/行）	传感器	分辨率/m
2000	2000-09-02	142/30	Landsat 5 TM	30
	2000-09-12	143/29
	2000-08-08	143/30
2010	2010-08-13	142/30	Landsat 5 TM	30
	2010-08-20	143/29
	2010-08-20	143/30
2020	2020-08-24	142/30	Landsat 8 OLI	30
	2020-07-14	143/29
	2020-09-16	143/30

判断依据	交互作用类型
q(X₁∩X₂)<Min[q(X₁),q(X₂)]	非线性减弱
Min[q(X₁),q(X₂)]<q(X₁∩X₂)<Max[q(X₁),q(X₂)]	单因子非线性减弱
q(X₁∩X₂)>Max[q(X₁),q(X₂)]	双因子增强
q(X₁∩X₂)=q(X₁)+q(X₂)	独立
q(X₁∩X₂)>q(X₁)+q(X₂)	非线性增强

土地利用类型	2000年		2010年		2020年
土地利用类型	面积/km²	比重/%	面积/km²	比重/%	面积/km²	比重/%
建设用地	241.43	1.70	542.73	3.82	877.86	6.18
绿地	5795.47	40.82	4818.91	33.94	5446.31	38.36
裸地	7728.13	54.43	8608.12	60.63	7568.68	53.31
水体	433.13	3.05	228.40	1.61	305.31	2.15
总计	14198.16	100.00	14198.16	100.00	14198.16	100.00

2000年	2020年
2000年	建设用地	裸地	绿地	水体	总计
建设用地	138.13	419.31	276.02	35.66	869.12
裸地	48.75	6168.61	1249.82	99.75	7566.93
绿地	38.82	1063.89	4252.87	103.17	5458.75
水体	4.37	92.32	15.51	191.16	303.36
总计	230.07	7744.13	5794.22	429.74	14198.16

因子	q值
因子	2000年	2010年	2020年
DEM	0.30	0.14	0.29
GDP	0.11	0.07	0.10
NDVI	0.34	0.35	0.29
河网密度	0.18	0.12	0.22
降水量	0.10	0.09	0.11
坡度	0.21	0.15	0.19
人口密度	0.12	0.02	0.06
日照时数	0.07	0.15	0.07
年平均气温	0.08	0.09	0.08