闽南三市城镇发展与地表温度的空间关系

基于厦门、漳州、泉州三市1996—2017年的Landsat遥感数据与夜间灯光数据,应用总体耦合态势模型及协调性模型探讨了城镇发展与地表温度的时空耦合规律,并运用标准差椭圆、双变量空间自相关及景观指数分析两者之间的空间响应规律,结果表明：1996—2017年,闽南三市表征城镇发展变化的夜间灯光与地表温度的空间分布呈现出与地形、区位等因素的空间耦合性。三市夜间灯光与地表温度的总体耦合态势不断加强,协调类型逐渐增多;城镇发展早期,夜间灯光对地表温度的影响存在滞后性,在城镇发展后期,夜间灯光则对地表温度的影响呈超前性;夜间灯光与地表温度呈正相关关系,空间溢出效应明显。夜间灯光对地表温度的正向影响逐渐加强,其与地表温度的HH、LL聚集不断扩张;HH集聚连片趋势明显。厦门的城镇发展对地表温度的影响更显著,而在泉州、漳州则较弱。本文可为闽南三市热环境优化提供科学指导。

Spatial relationship of urban development to land surface temperature in three cities of southern Fujian

Exploring the influence mechanism of expansion for urban areas on thermal environment is significant for improving urban ecological environment. In this study, nighttime light(NTL) can be regarded as an evidence of urban development. Based on Landsat remote sensing data, DMSP/OLS nighttime light data and NPP/VIIRS nighttime light data of Xiamen,Zhangzhou and Quanzhou in southern Fujian during 1996-2017, this study applied the overall coupling model and coordination model to discuss the spatio-temporal coupling and coordination relationship between urban development and land surface temperature(LST)distribution. Additionally, the spatial response law was analyzed by standard deviation ellipse,bivariate spatial autocorrelation and landscape index. The results show that during 1996-2017,the spatial distribution patterns of LST and NTL show that urban development is closely related with factors such as geographical location and terrain. In the western inland areas with high elevations, forests are obviously concentrated with relatively low LST, while in the eastern plain areas, more urban areas are distributed with relatively high LST. In the three cities, the overall coupling situation of NTL and LST is constantly strengthening, and the proportion of coordination for NTL and LST is gradually increasing. In the early stage of urban development,the influence of NTL on LST is hysteretic. In the late stage, the influence of NTL on LST is in advance. There is a positive correlation between NTL and LST, and a spatial spillover effect is obvious. The correlation coefficient and bivariate spatial autocorrelation Moran’s I value gradually increase, indicating that NTL has an increasing influence on the change of LST. The HH(High-High)-type and LL(Low-Low)-type agglomeration areas continue to expand.Influenced by the trend of integrated development of the three study cities, HH-type agglomeration area is gradually concentrated in regions connecting the cities, such as Xiamen,Jinjiang, Shishi and central urban area of Zhangzhou. The influence of urban development on LST is related to the development condition of itself. Compared with Zhangzhou and Quanzhou, NTL has a more significant influence on the LST in Xiamen. The study results provide a scientific guidance for the optimization of thermal environment in the three cities of southern Fujian.