山地城市土地覆盖变化对地表温度的影响

Influence of land cover change on land surface temperature in the mountainous city

针对山地城市复杂的城市地貌和下垫面类型,本文使用TM、DEM、ETM＋等遥感影像资料,提取了重庆市土地利用覆盖类型;借助TM、MSS等遥感数据的红外波段,反演出1988年和2000年的地表温度。分析了重庆市近十年的土地覆盖变化及其对地表温度的影响,结果表明,在1988-2000年间,研究区土地覆盖变化明显,特别是城市土地覆盖面积有显著增加。土地覆盖类型的变化会改变地表温度的空间分布,尤其是城市土地的扩展会提高地表温度。对山地、丘陵、平坝、陡坡四种耕地的地表温度进行了深入分析与研究,结果表明：山地城市土地覆盖变化引起了植被覆盖度的变化,而植被覆盖度的变化又相应地影响了地表温度的变化,植被覆盖度每下降10%,地表温度上升0.49K。

Based on complex physiognomy of the mountainous city, as well as its underlying surface type, this paper used TM, DEM and ETM＋ data to extract the land-use cover types of Chongqing, and retrieved the land surface temperature by using single Landsat TM/ETM6 thermal infrared data to analyze complex urban landscape of 1988 and 2000. The integration of remote sensing and GIS was further applied to examine the influence of land cover change and urban growth on land surface temperatures. The article analyzes land-cover change in Chongqing during recent 10 years and its influence on land surface temperature. The results show that the combination of various land cover types extracted by effective infrared band and retrieved land surface temperature space as well as plantcovered space, could accurately reveal the relationship of plant-covered degree and retrieved land surface temperature. And then this paper demonstrates that the rise of land surface temperature in Chongqing is mainly due to urban land development. In the period 1988-2000, the land-cover of the study area changed significantly, in particular, the urban construction land area increased significantly. The land-cover change led to spatial changes of the LST distribution. Furthermore, urban development in Chongqing raised land surface radiant temperature. We studied four types of cultivated land surface temperatures related to mountains, hills, fattened land （intermontane basins） and steep slopes. The results show that the land-cover changes in the mountainous city caused the changes of NDVI, and the change of NDVI caused the land surface temperature. As NDVI decreased by 10%, the land surface temperature rose by 0.49K. Finally the essay concludes that the increase of accuracy defined by land surface emissivity could increase the accuracy of retrieved land surface temperature. The increase of categorical accuracy in land use types could demonstrate how land use change influences land surface temperature in the mountainous city, whichshould be emphasized in further researches.