| 摘要: |
| 面对快速城市化带来日益严重的热环
境问题，充分发挥蓝绿空间的生态降温作用是
实现人居环境治理及可持续发展的重要路径。
本文以天津市内六区为研究区，应用遥感（RS）
与地理信息系统（GIS）技术开展蓝绿空间与热
环境的关联讨论。首先，基于Landsat遥感影像
进行地表温度反演及蓝绿空间解译，分析比较
热环境与蓝绿空间的分布特征。其次，运用统计
分析方法探讨蓝绿空间景观格局指数与热环境
的相关规律后发现：在斑块层面，蓝绿空间内部
温度与斑块周长呈显著负相关，与周长面积比
呈显著正相关；在类型层面，蓝色空间样本研究
最佳尺度为1 500 m×1 500 m，样本温度与斑
块所占景观面积比、最大斑块指数、平均斑块面
积、边缘密度、斑块密度呈显著负相关；绿色空
间最佳尺度为300 m×300 m，样本温度与斑块
所占景观面积比、最大斑块指数、平均斑块面积呈显著负相关。最后，在规划管理与设计层面提出蓝绿空间规划建议，以期为热环境改善视角下
的城市蓝绿空间规划提供参考。 |
| 关键词: 蓝绿空间 热环境 地表温度 景观格局 |
| DOI：10.13791/j.cnki.hsfwest.20230216 |
| 分类号: |
| 基金项目:国家自然科学基金面上项目（52078329）；国
家自然基金委国际（地区）合作与交流项目（52
061160366）；西南科技大学自然科学基金项目（22
zx7158） |
|
| Study on the Relationship Between Landscape Pattern of Urban Green and Blue Space andThermal Environment of Tianjin |
|
TAN Ning,CHEN Tian,LI Yangli
|
| Abstract: |
| Urban thermal environment refers to the physical environment system affecting human
survival and development. Nowadays, high temperature, heat island effect and other problems are
common in major cities in China, which not only affect the quality of living environment, but also
seriously threaten the sustainable development of cities. At present, most researches focus on green
space such as urban parks, relatively weakening or ignoring the cooling effect of water bodies. And
there are few systematic researches that focus on green and water space at the same time. It is of
great significance to identify the correlation between urban thermal environment and landscape
pattern of green and blue space to fully realize the cooling effect.
Tianjin is located in the northeast region of the North China Plain and typically has a warm,
temperate and sub-humid monsoon climate. In this research, six districts of Tianjin were chosen
to form the research area. In recent years, the six districts of Tianjin are the strongest areas of
Urban Heat Island Intensity (UHII ), which expand from the strong heat island in the city center to
the surrounding areas, with the annual average UHII in the urban area reaching more than 1 ℃.
However, there are few discussions about thermal environment of Tianjin, which is urgent to solve
the thermal environment problem.
Firstly, the Landsat 8 remote sensing image covering the research area with little cloud was
selected as the data source, and the Remote Sensing technology was used for preprocessing. The
surface temperature distribution maps of seven thermodynamic hierarchies were obtained by using
the Mono-window Algorithm with high precision. Next, the landscape types of the research area
were divided into three categories through the Supervised Classification: blue space including
natural and artificial water bodies, green space including natural and artificial green space, and non-
blue-green space refers to urban construction land other than blue-green space. Then three indicators
(AREA, PERIM, PARA) were selected at patch level, and 10 indicators (PLAND, LPI, AREA_MN,
ED, SHAPE_MN, LSI, FRAC_AM, AI, PD, DIVISION) were selected at class level for calculation
and statistics. Finally, the correlation between landscape pattern index of urban blue and green space
and the thermal environment at patch level and class level was analyzed, and the regression model
was established for significance test. Because the spatial pattern of different scales will lead to the
loss or change of landscape information characteristics, the moving window method was used to
discuss the thermal environment response of multi-scale sub-landscape samples. Combined with
the window scale used in previous research on related issues, and the pixel scale ( 30 m × 30 m ) of
Landsat remote sensing images, it was decided to use five rectangular grids of 300 m × 300 m, 600 m
× 600 m, 900 m × 900 m, 1 200 m × 1 200 m, and 1 500 m × 1 500 m. The research area was divided
into 1 844, 421, 177, 90 and 54 sub-landscape samples to study the correlation between urban blue
and green space and surface temperature.
The conclusions of the research are as follows：1) The distribution of thermal environment
is closely related to urban blue and green space. The urban low temperature range is mainlyconcentrated in blue and green space such as rivers, lakes and parks, which characterizes its significant thermal environment improvement effect. In the
cooling effect, the blue space is slightly stronger than the green space. 2) At patch level, the internal temperature of blue and green space is significantly
negatively correlated with the PERIM and significantly positively correlated with PARA. 3) At class level, the optimal scale of blue space sample study is 1
500 m × 1 500 m, and the sample temperature is significantly negatively correlated with PLAND, LPI, AREA_MN, ED and PD. The optimal scale of green
space is 300 m × 300 m, and the sample temperature is significantly negatively correlated with PLAND, LPI and AREA_MN.
According to the above findings, this paper put forward relevant suggestions on blue and green space planning at the level of planning management
and design. Based on the empirical study, the optimization scheme of blue and green space structure in six districts of Tianjin was put forward to form an
ecological system of “four axes, one ring, multiple corridors and multiple nodes”, so as to provide references for urban blue and green space planning from
the perspective of thermal environment improvement. Future researches in this area should select higher resolution data sources to propose more accurate
and scientific optimization strategies. |
| Key words: Green and Blue Space Thermal Environment Land Surface Temperature Landscape Pattern |
