沿海河谷盆地城市热岛时空特征及驱动机制

Spatial-temporal characteristics of urban heat islands and driving mechanisms in a coastal valley-basin city: a case study of Fuzhou City

针对以往热岛研究中缺少自然-人文多维因子组合作用的成因分析,基于landsat TM/OLI遥感影像反演福州市1991、2000及2013年河谷盆地亮温热场,并利用主成分多元回归方法探讨DMSP/OLS夜间灯光、交通密度、地表非渗透面、地面高程、地表植被、水体等多维因素对城市热岛的综合影响及联动关系。研究发现:11991—2013年盆地热岛辐射范围不断蔓延,老城区成为热岛核心区,呈中心向外辐射并沿江扩展态势,盆地"溢出"现象显著;2盆地内平原-山地温差达10—14℃,热岛面积出现两极分化,23年间热岛面积共增加226.9km^2,其中高温区面积增长四倍,年增长率最大(7.50%);3夜间灯光、交通路网密度、地表非渗透面、地表植被、地面高程及水体每变化1个单位,温度将分别变化0.430、0.418、0.103、-0.031、-0.469及-0.096;4人文促进作用大于自然抑制作用,综合作用下盆地升温0.35个单位;5相比地表非渗透面,夜间灯光与交通密度对热岛的增温贡献更为显著;地面高程主导降温作用,其次为水体与地表植被,后两者作用之和仅为地面高程的27%。福州盆地当前的生态环境并不足以缓解城市未来进一步发展而导致的热岛加剧,当保持城市现有发展规模且自然作用进一步增加60%时,降温抑制作用才会明显。

Previous literature on cause analyses of urban heat islands （UHI） lacks perspective from the combined effects of natural and humanity factors. With this prerequisite, the present study used Landsat TM/OLI remote sensing images to extract urban thermal fields of Fuzhou city, a typical valley basin in China, during 1991, 2000, and 2013. The combined effect and linkage relationship between the multidimensional factor and UHI was also explored using DMSP/OLS, traffic density, and land surface information indexes, including normalized difference building index （NDBI）, digital elevation model （DEM）, normalized difference vegetation index （NDVI）, and modified normalized difference water index （MNDWI）, based on GIS and principal component multiple regression. The results revealed that （1） the UHI of the Fuzhou valley basin has spread outward along the Minjiang River while the inner-city has become the core area of the UHI, commencing its significant overflow heat phenomenon from the center to outer edge from 1991; （2） the temperature difference between the plain and mountainous areas in the basin reached 10-14℃, and the UHI area displayed a polarization appearance indicating that both the low and high temperature areas have tendency to increase. Moreover, the high temperature area increased four times to 138.5 km2 between 1991 and 2013, with an average annual growth rate of 7.5%; 3） according to the coefficient simulation, the basin temperature changed 0.430, 0.418, 0.103, -0.0310, -0.469, and -0.0963 as the night light intensity, NDBI, traffic density, NDVI, DEM, and MNDWI increased every 1 unit respectively; （4） the human promoting effect was greater than the natural inhibition on the UHI, resulting in the basin warming by 0.35 units; and （5） in comparison to the NDBI, night light intensity and traffic density contributed more significantly to the warming of the UHI, and the cooling effect of surface elevation is prominent, followed by water and surface vegetation, with these accounting for only 27% of the surface elevation effect. In addition, the cooling effect was obvious when the natural effect further increased by 60% and Fuzhou city simultaneously remains at the existing developmental scale. This also reveals that the current ecological environment of the Fuzhou basin is not enough to alleviate the UHI caused by future development of the city.