【作者】 历华；

【导师】 黄健柏； 曾永年；

【作者基本信息】 中南大学 ， 地图学与地理信息系统， 2007， 硕士

【副题名】以长株潭地区为例

【摘要】 近年来，人们对城市环境的关注使城市热岛研究得到越来越多地重视，如何定量监测、分析和评价城市热岛效应，已成为当前城市气候与环境研究重要的研究内容之一，也是全球变化研究的重要方面。同时随着遥感技术的发展，城市热岛研究也成为定量遥感应用的重要领域。因此，开展城市热岛效应的遥感研究具有十分重要的理论与实践意义。本文利用多源遥感数据的优势，以长株潭地区为例，就地表温度反演方法、基于地表温度的区域和局地尺度城市热岛的应用研究进行了深入的研究，主要研究内容及结论如下：(1)基于多源遥感数据的地表温度反演方法研究针对目前利用单通道热红外数据反演地表温度时大气参数的难以获取，提出利用多源遥感数据的优势来反演地表温度的方法。首先利用MODIS近红外数据在像元尺度上获取温度反演中所需的大气参数——大气水分含量，再利用同时相的Landsat TM／ETM+数据，采用普适性单通道算法反演地表温度。研究结果表明，多源遥感数据的综合应用是城市地表温度反演的有效途径与方法，可获得较合理的地表温度反演结果。这种方法的最大优点就是可以快速、精确获取地表温度反演中所需要的大气参数。(2)区域尺度的长株潭城市热岛研究基于多季相MODIS影像，采用分裂窗算法反演地表温度，对长株潭地区城市热岛空间分布与季相变化特征、影响因子进行定量研究。结果表明，长株潭地区春季和夏季存在明显的城市热岛效应，而冬季和秋季城市热岛并不明显；地表覆盖类型对城市热岛的影响十分明显，长株潭地区春、夏、秋季植被绿地状况与城市热岛呈现明显负相关分布，其中以夏季最为明显，夏季地表温度与NDVI的复相关系数R~2达到0.8193，即植被覆盖对城市地表温度的影响显著。因此，城市植被的分布与季节变化影响着城市热岛的强度与时空分布，揭示出植被绿地对降低城市热岛效应具有重要的作用，大范围的绿地建设能有效降低城市热岛效应。(3)局地尺度的长沙市城市热岛研究基于多时相Landsat TM／ETM+影像，采用单通道算法反演长沙市地表亮温；利用NDVI(归一化植被指数)，MNDWI(改进的归一化水体指数)，NDBI(归一化建筑指数)和NDBaI(归一化裸土指数)4个指数，采用决策树分类方法，对长沙市土地利用／覆盖进行的分类。在此基础上，对长沙市城市热岛的空间分布特征、时空演变特征以及城市热岛与土地利用／覆盖变化和各种影响因子之间的关系进行定量研究。结果表明，随着长沙市城区范围的不断扩张，城市热岛范围也不断增大，土地利用／覆盖类型的变化会改变地表温度的空间分布，城市用地和裸地对地表温度的贡献最大，是城市热岛强度的主要贡献因素，而水体和林地具有较好的降温作用。地表温度与4种指数的回归分析表明地表温度与这些指数之间有很明显的相关性，不同土地利用／覆盖类型的地表温度存在较大差异。更多还原

【Abstract】 With the increasing of urban environment issuses, scientists have paid more attention on Urban Heat Island(UHI)research. Therefore, how to monitor, analyze, and evaluate UHI effect quantitatively has become one of the prime impormant research aomong current urban climate and environment research. With the development of remote sesing technology, it has become an important approach to UHI researches. In this paper, combined with the multi-source remote sensing data(MODIS and Landsat TM/ETM data), the improvement on the method for land surface temperature(LST)retrieval from single thermal remote sensing data have made successfully. Taking Changsha-Zhuzhou-Xiangtan metropolis area as a case study area, the author studied UHI effect at regional and local level. The main conclusions are as follows:(1) Retrieval of LST with multi-source remote Sensing DataIn order to get the atmospheric parameter of LST retrieval by single channel thermal infrared data, this paper calculated firstly the distribution of atmospheric water vapor content at pixel level using MODIS near infrared data. Then, LST is retrieved by using the thermal infrared band of Landsat TM/ETM+ and atmospheric water vapor content based on generalized single-channel method. The exploited approach was applied to Changsha and Wuhan city. The results show that the combination of multi-source remote sensing data is effective for retrieving LST and estimates reasonably land surface temperature. The most advantage of this approach is to obtain the atmospheric parameter of LST retrieval rapidly and accurately.(2) UHI study of Changsha-Zhuzhou-Xiangtan area at regional levelLST in Changsha-Zhuzhou-Xiangtan area was retrieved by split window algorithm based on multi-season MODIS data. The seasonal characteristics of UHI and its factors were analyzed quantitatively in Changsha-Zhuzhou-Xiangtan area. The result indicated that there was a clear UHI effect in the spring and summer, but this did not happen in the winter and autumn; land cover affected urban heat island clearly, the distribution of UHI presented a obvious negative correlation with the vegetation in spring, summer and autumn in Changsha-Zhuzhou-Xiangtan area, especially in summer, the correlation coefficient R~2 between LST and NDVI has reached 0.8193 in summer, that is, the higher vegetation cover, the less urban heat island was. Therefore, the vegetation has an important effect on alleviating urban heat island.(3) UHI study of Changsha city at local levelFirstly, LST was retrieved by single channel algorithm based on multi-temporal Landsat TM/ETM+ data. Then NDVI, MNDWI, NDBI and NDBaI indices were calculated to classify the land use/cover types by decision tree method in the study region. Finally, the spatial and temporal changes of UHI and the relationship between UHI and land use/cover change in Changsha city was studied quantitatively. The results show that with the expansion of built-up area in Changsha city, the extent of UHI was increasing. The change of land use/cover types would change the spatial distribution of LST. Bulit-up area and bare land were the maijor contribution factors of UHI intension, while water body and froest land has a good function of reducing LST. The regression analysis shows that there were significant correlations between LST and four indices, and there was a great difference of LST in different land use/cover types.更多还原