香港城市热岛效应的遥感反演分析

A Remote Sensing Study of the Urban Heat Island of Hong Kong

摘要: 城市热岛是当今社会的热点问题，被认为是主导城市环境和气候的要素之一。本文将利用2005年的Landsat TM影像对香港的城市热岛效应进行研究。通过覃志豪的单窗算法反演地表温度，进而分析香港地区地表温度的空间分布特征以及由此引起的热岛效应；并利用地表温度与NDVI、NDBI指数的相关性分析，探讨城市绿地和建筑用地对热岛效应的影响，为城市的生态环境建设提供依据。结果表明，香港的城市热岛效应主要集中于香港国际机场和九龙半岛附近。香港国际机场自1998年建成以来，周围建筑用地急剧增加，导致了热岛效应；而在商贸集中的九龙地区，由于建筑物大规模的连绵和扩展，造成了区域大面积的温度升高，城市热岛甚至发展成为区域热岛。同时，地表温度与NDVI、NDBI指数的相关性分析显示，城市绿地可有效的缓解建筑用地增加所导致的城市热岛效应，因此在以后的城市规划和发展过程中，应当重视城市绿化，积极增加绿地覆盖度。

Abstract: As a hotspot issue of urban environment and climate, urban heat island effect has aroused wide public concern. In this paper, based on Landsat TM data of Hong Kong (HK) in 2005, a research of urban heat island effect will be implemented. The mono-window algorithm which is proposed by Qin and only requires 3 essential parameters without the use of in situ measurements is used to retrieve the land surface temperature (LST), and we give a detailed description of the algorithm in the paper. Then next, we could analyze the spatial pattern of LST in HK and get the distribution characteristics of urban heat island effect. On the other hand, we also do the correlation analysis between LST and NDVI, NDBI to explore the impacts of green land and build-up land on urban heat island and provide useful information for the urban development and environment protection. The results show that, the Hong Kong urban heat island effect is mainly located in Hong Kong international airport and Kowloon Island. Since the start of Hong Kong international airport in 1998, the construction land increases sharply. The change of land cover/use has conducted to the urban heat island effect in Chek Lao Kok. In the Kowloon area, the urban heat island has converted to regional heat island because the connected pieces of urban construction. Otherwise, the 9 dispersed satellite cities of HK have dispersed distribution of urban heat island effect. So in city layout, the establishment of satellite cities apart from the business center is reasonable to prevent the formation of large-scale regional urban heat island. Meanwhile, we get the correlation coefficient between LST and NDVI, NDBI through the correlation analysis. The negative correlation coefficient of LST and NDVI proves that the green land can weaken urban heat island effect; while the positive correlation coefficient of LST and NDBI demonstrates that the build-up land can contribute to the effect. Thus, in our future city planning and development, we should pay attention to the urban greening.