城乡不透水面增长格局及地表温度的响应特征研究

Pattern of Impervious Surface Change and Its Effect on Land Surface Temperature in Urban and Rural Areas

为探究中国北方中温带,特别是东北寒区快速城市化地区城乡不透水增长格局及地表温度的响应特征,本文以哈尔滨市为例,基于国家资源环境遥感时空信息平台土地利用/覆盖变化(LUCC)数据集解译的2001年与2015年城乡建设用地和Landsat 7/8数字遥感影像,结合植被-不透水面-土壤(V-I-S)端元选取和完全约束最小二乘混合像元线性分解模型进行了不透水面提取(分辨率15 m×15 m),并运用单窗算法进行了夏季地表温度遥感反演。结果表明:2001-2015年建设用地扩张259.05 km2,不透水面上升163.96 km2,城市与乡村不透水面占各自建设用地的比例由2001年的43.92%、21.35%变化为2015年的49.14%、34.27%,城乡比例差由22.57%缩减至14.87%,单位建设用地内乡村不透水面增量较高;2001-2015年城区以低温区、中温区、高温区为主,对不透水面扩张的响应剧烈,而乡村以低温区和中温区为主,低温区和高温区响应剧烈;地表温度与不透水面具有显著正相关,在低、中、高不透水密度区分别升温1.16o、1.45和1.79℃,相同不透水面盖度下城市升温高于乡村。总体而言,研究区不透水面大幅扩张,温度分区变化剧烈,地表温度随不透水面增加升温效果明显。

As a big city in Northeast China, Harbin has experienced dramatic urban expansion in recent decades, followed with increased impervious surface area （ISA） and land surface temperature （LST） effects. In order to explore the relationship between ISA and LST under such a rapid urbanization process in urban and rural areas, based on the National Environmental Remote Sensing temporal resource information platform of land use/cover change （LUCC） data set, we extracted the construction land for urban and rural areas between 2001 and 2015. Combined with the end-member selection model for vegetation-impervious surface-soil （V-I-S） and the fully con- strained least squares linear mixed pixel decomposition model, we extracted the impervious surface （a resolution of 15 m~ 15 m）. The mono-window algorithm and Landsat images were used to retrieve the LST of Harbin City in the summer of 2001 and 2015. The results showed that： （1） ISA had increased by 163.96 km2 and an expan- sion of 259.05 km2 had occurred to the construction land, indicating an increase of 10.26% between 2001 and 2015; the proportions of ISA in construction land for urban and rural areas are 43,92% and 21.35% respectively in 2001, and 49.14% and 34.27% in 2015. The ratio of the poor had reduced from 22.57 to 14.87%, and the change of per unit construction land in the rural area is more sensitive. （2） The urban areas are dominated by inferior low temperature, medium temperature and inferior high temperature, with fast increasing speeds; while the rural areas are dominated by inferior low temperature and medium temperature, within which the inferior low temperature and inferior high temperature have faster increasing speeds. （3） LST and ISA have a significant positive correlation, and the temperatures for the low, medium and high-density areas of ISA were raised up by 1.16, 1.45 and 1.79℃ respectively. Also, the temperature rose faster in the urban area than the rural area under the same coverage of ISA. In summary, the partitions of LST have severe changes, which are accompanied with substantial expansion of ISA, and the increase of LST is significant in response to the expansion of ISA.