随着城市化进程的加剧，城市景观配置发生显著变化，地表自然景观逐渐被不透水面等人为景观所替代，驱动地表温度发生变化，导致区域气候改变。以往单一城市不透水面和地表温度关系的研究主要在单一城市尺度开展，但是由于单个城市到城市群的空间异质性以及城市群之间气候背景相关差异的影响，不透水面的增温效应的时空尺度独立性仍需进一步研究。此外，综合不透水面、绿地空空间等城市地表组分对城市化区地表温度的影响虽已有较多研究，但对于城市景观组分及配置对城市化区域地表温度的影响则涉及较少。且已有研究多以某一城市或都市区为研究案例，较少涉及不同气候背景(不同纬度)下的城市群内，不同规模等级城市地表组分空间模式对地表温度时空异质性研究。 因此，本研究以中国东部主要的三大城市群京津冀、长三角和珠三角为研究案例，基于 Landsat 遥感影像数据和 MODIS 地表温度数据， 初步探究了不同气候背景下以及不同空间尺度，不透水面对地表温度增温效应的时空差异性；进而整合不透水面、城市绿地空间、水体等城市地表组分，同时考虑地表组分空间格局及配置，量化三大城市群区域不同规模等级城市尺度地表组分空间格局对地表温度时空异质性的影响， 并在此基础上，模拟了未来地表组分和地表温度的空间演变模式。

       研究发现： (1)不透水面对地表温度的增温效应存在昼夜和季节的时间差异；空间尺度上，单一城市尺度不透水面对地表温度的增温效应比城市群尺度更加显著。其中单一城市不透水面和地表温度的相关系数可达到 0.94，说明在城市尺度，不透水面可以更好的指示地表温度的变化。 (2)不透水面和地表温度的关系不能仅用线性关系来描述，受到空间尺度、纬度位置和气候背景的影响， 京津冀城市群不透水面密度高于 10%-17%时， 呈现分段线性关系。此外，城市群间的结果对比显示，珠三角城市群不透水面对地表温度的增温效应强于京津冀和长三角。 (3) 2000-2015 年三大城市群内各个规模等级城市地表温度普遍升高，不透水面逐年扩张且城市绿地空间占比逐年降低。不透水面地表温度高于城市绿地地表温度约 0.96-7.96℃。 (4)通过量化地表温度与不透水面和城市绿地空间组成密度以及空间形状配置的关系，我们发现城市规模等级相对越高的城市，不透水面对地表温度影响更强。地表组分空间格局对地表温度影响研究表明，地表组分的组成和配置对地表温度影响的综合效应最强，且与地表组分的组成及配置单独对地表温度的影响存在量级差异。此外，不透水面和绿地空间组成对地表温度的影响普遍强于其空间配置的影响。 (5)各个规模等级城市未来不透水面将持续扩张，城市绿地所占比例持续缩小，地表温度也持续升高。规模等级越高的城市，不透水面和绿地空间的地表温度差越明显。

      With the intensification of urbanization, the urban landscape configuration has undergone significant changes, and the natural surface landscapes are gradually replaced by man-made landscapes such as impervious surface area, which driving changes in land surface temperature and resulting in regional climate changes. In the past, there are many studies focus on the individual city about the relationship between impervious surface and land surface temperature. However, due to the spatial heterogeneity of individual city to urban agglomerations and the impact of differences in climate background between urban agglomerations in different geographical locations, the temporal and spatial heterogeneity impact of impervious surface on land surface temperature still needs further study. Moreover, there are many studies on the impact of urban land surface components such as integrated impervious surface area and urban green space on the land surface temperature over urbanized areas. However, there have been fewer studies on the impact of urban landscape components and configurations on the surface temperature of urbanized areas. And most of the existing studies take a certain city or metropolitan area as a case study, and seldom involve the spatial and temporal heterogeneity influence of the land surface temperature of different scale and level cities in urban agglomerations under different climate backgrounds (different latitudes). Therefore, this study takes the three typical urban agglomerations in eastern China as research cases. Based on Landsat remote sensing data and MODIS land surface temperature data, preliminarily explored the temporal and spatial differences of the warming effect of impervious surface on land surface temperature under different climate backgrounds and different spatial scales. Furthermore, this study integrates urban land surface spatial components such as impervious surface area, urban green space, water bodies, and considers the spatial pattern and configuration of the land surface components, quantifies the spatial and temporal heterogeneity impacts of the spatial distribution pattern of land surface components on land surface temperature at different scales and levels cities. And on this basis, the future spatial evolution model of land surface spatial components and land surface temperature was simulated.
       The main results of this study are as follows: (1) The warming effect of impervious surface on the land surface temperature differs in time scales including diurnal and seasonal scales. At the spatial scale, the warming effect of impervious surface on the land surface temperature on an individual city scale is more significant than that on the urban agglomerations scale. Among them, the correlation coefficient between impervious surface and land surface temperature at the individual city can reach 0.94, indicating that at the individual city scale, the land surface temperature is more significantly affected by the impervious surface. (2) The impervious surface and land surface temperature relations cannot be described by linear relationship alone. Because of the spatial scale, latitude position and climate background, when the impervious surface density is higher than 10%-17%, we depicted piecewise linear relationship between impervious surface and land surface temperature in Beijing-Tianjin-Hebei urban agglomeration. In addition, among the three major urban agglomerations, the warming effect of impervious surface on land surface temperature is more significant than the other two. (3) We found generally rising land surface temperature over all cities considered in this study during 2000-2015 and which can be attributed to continual urban expansion-induced decreased urban green space. Generally, land surface temperature over impervious surface area is 0.96-7.96℃ higher than that over urban green space. (4) We investigated impacts of spatial configuration of the surface components on land surface temperature and found the most significant are the combined impacts of surface components and relevant spatial configuration on land surface temperature. Specifically, the impervious surface temperature and urban green space have higher impact on land surface temperature than the geometry of the impervious surface temperature and urban green space on land surface temperature. (5) In the future, continual expansion of impervious surface area and continual shrinking of urban green space will drive rising tendency of land surface temperature over all cities. Moreover, larger rising tendency of land surface temperature will be observed in larger-size city than over smaller-size city.