地理学报 ›› 2016, Vol. 71 ›› Issue (12): 2170-2184.doi: 10.11821/dlxb201612008

• • 上一篇    下一篇


张甜(), 王仰麟(), 刘焱序, 彭建   

  1. 北京大学城市与环境学院 地表过程分析与模拟教育部重点说实验室,北京 100871
  • 收稿日期:2015-12-01 修回日期:2016-07-22 出版日期:2016-12-25 发布日期:2016-12-26
  • 作者简介:

    作者简介:张甜(1991-), 女, 陕西汉中人, 博士生, 研究方向为综合自然地理与景观生态。E-mail: zhangtiangis@163.com

  • 基金资助:

Multi-temporal detection of landscape evolution in western Shenzhen City during 1987-2015

Tian ZHANG(), Yanglin WANG(), Yanxu LIU, Jian PENG   

  1. Laboratory for Earth Surface Processes, Ministry of Education, College of Urban and Environmental Sciences, Peking University, Beijing 100871, China
  • Received:2015-12-01 Revised:2016-07-22 Online:2016-12-25 Published:2016-12-26
  • Supported by:
    National Natural Science Foundation of China, No.41330747



关键词: 城市扩张, 生态响应, 拐点识别, 分段线性回归, 深圳市


The expanding impervious surface along with the process of urbanization has produced great impact on ecological land in most of megalopolises all over the world. A typical urban expansion process usually shows a phase character, and the rapid increase of urban area has a non-linear correlation with vegetation degradation. Therefore, it is essential to recognize the important turning points in the curve of urban growth and the expansion pattern, which would also make a clear understanding of ecosystems' responses to urbanization. Landsat TM/ETM+ dataset was widely used in the detection of landscape change, however, most of the studies used the visual interpreted data per 5 years or above to characterize a long-term variation, and less attention was paid to the identifications of urban sprawl stages. In this study, with western Shenzhen City as a case study area, two new indexes of MNDBI* and NDVI* were built based on long-term Landsat dataset during 1987-2015, in order to identify the turning points of urban sprawl in temporal dimension and analyze the landscape evolution characteristics at different expansion axes. The results showed that western Shenzhen had experienced a rapid urbanization and the water area as well as vegetation coverage decreased dramatically. Specifically, the urban growth could be divided into two phases, i.e. rapadly growing period and stable period by the year 2003, and the vegetation degradation changed in 1999. The similar turning points confirmed the consistent spatiotemporal correlation between urbanization and the ecosystem responses. Furthermore, from the perspective of spatial differentiation, urban sprawl consistently occurred in the northern districts and coastland with sea reclamation until 2003-2006. The southern districts in western Shenzhen experienced the stable urbanization period much earlier than the other ones, with the turning points were identified during 1995-1998. Urban landscape evolution showed various characteristics in different profiles, a large number of sample points were selected through the west, north, east, northwest, northeast directions per 120 m, and the two indexes were calculated in 1987 and 2015 in four directions from central Shenzhen. Obvious urbanization as well as vegetation degradation could be found in a further expansion radius. Also, it is indicated that the north and west profiles should be the major urban expansion axes. In a word, Shenzhen's urbanization has exerted a complex effect on the eco-environment, and in order to realize vegetation restoration, a series of environmental protection policies were promulgated and implemented in recent 10 years. However, this could only postpone the process of habitat degradation rather than ecological restoration. Therefore, more powerful policies are required in the future development of Shenzhen City, and more attention should be focused on the response of ecosystems to the urbanization process.

Key words: urban sprawl, ecological response, turning point identification, piecewise linear regression, Shenzhen City