【作者】 李珊珊；

【导师】 宫辉力；

【作者基本信息】 首都师范大学 ， 地图学与地理信息系统， 2009， 硕士

【摘要】 采用传统的气象资料不能实时、动态、连续的反映地表温度空间分布的差异。卫星遥感为地表温度的求取提供了新思路,利用MODIS ( Moderate Resolution Imaging Spectroradiometer,中分辨率成像光谱仪)数据宏观、实时、周期性的特点,可以很好地对地表温度进行分析。随着遥感应用的深入,现有的热红外遥感数据地表温度反演算法分为三大类:单通道算法、分裂窗算法和多波段算法。本文比较了这三种算法,并建立了基于MODIS和分裂窗算法的北京市地表温度反演模型。其中,用来演算地表温度的分裂窗算法是以NOAA--AVHRR所观测到的热辐射数据为基础。根据Planck热辐射函数,将AVHRR的两个热通道(即通道4和5)数据转化为相应的亮度温度。选择与AVHRR的两个热通道(4,10.5-11.3μm和5,11.5-12.5μm)波段范围接近的MODIS 31(10. 780- 11. 280μm),32(11. 770- 12. 270μm)两波段来计算。通过对Planck方程的辐射强度和温度计算模拟,发现温度和辐射强度的近似关系,由此建立线性方程,以简化计算。然后通过模拟计算得到大气透过率与大气水汽含量的关系,从相同MODIS影像数据中计算得到MODIS 31/32波段透过率。最后运用分裂窗算法和MODIS的31和32通道反演地表温度。研究表明:利用MODIS数据和分裂窗算法反演北京市地表温度是可行的。结果发现:随着北京城市建设和城市化速度的加快,城市地表热量分布有着明显的季节变化分布特征。具体如下:春季,由于地表植被覆盖低,城市热岛强度很低、近似为0;夏季热岛强度和热岛面积都增至最大,热岛面积甚至超过五环,到达通州;秋季热岛强度和面积减小;冬季有冷岛现象存在,即城市地表温度低于乡村。揭示出植被覆盖对城市地表温度的影响显著。城市植被的分布及季节变化影响着城市热岛的强度与时空分布,对于降低热岛效应有显著效果。大气气溶胶是指由自然过程和人类活动造成,悬浮在大气中沉降速度小、尺度范围为0.001-20μm大小的液态或固态粒子。气溶胶对天气和气候起着重要作用,同时对人类的居住环境产生巨大的影响。遥感具有面状观测、准实时获取、更新周期短、成本低的特点,对城市地区大气污染的重要组成部分气溶胶,可以进行快速的监测。本文利用MODIS 1B数据为数据源,使用6S进行辐射传输计算构建查找表,进行气溶胶光学厚度的反演,并对其结果进行了比较。结果表明,该算法在假定气溶胶与大气分子的散射相函数相同的情况下,能较好得监测气溶胶,反映城市气溶胶的区域变化,结合北京市及周边地区的地理状况进行分析,具体如下:气溶胶光学厚度的高值区基本分布在人口密集、交通和工商业活动频繁的地区,高分辨率卫星遥感结果对污染物的排放源分布监测具有潜在的应用价值,能够在一定程度上表征北京市空气质量的状况。通过对北京市地表温度和气溶胶光学厚度在空间水平尺度上的对比,以夏季为例,可以发现北京市地表温度和气溶胶光学厚度高值区虽然大致都是北京城区即人口密集、交通和商业活动频繁的地方,但北京城市热岛效应和空气污染主要来源于城市本身,温度对气溶胶浓度变化的影响并不大。由于温度在一定程度上可以表征城市热环境,气溶胶光学厚度在经过垂直和湿度影响因子订正后,可以作为监测颗粒物污染物地面分布的一个有效手段,用以表征城市污染,而可吸入颗粒物是目前北京首要空气污染物,在此基础上,本文结合北京市环保局发布的空气质量状况和地面实测数据,从微观上对温度和可吸入颗粒物进行分析,发现城市环境中温度与可吸入颗粒物个数关联影响不大。从温度的变化来看,可吸入颗粒物的个数随着温度的升高反而减少,也就是说,温度高,可吸入颗粒物个数反而少,这可能是高温时地面湍流强烈,大气垂直方向输送较大,垂直方向有利于可吸入颗粒物的扩散,所以空气中的可吸入颗粒物个数偏少。更多还原

【Abstract】 With the acceleration of urbanization, the phenomenon of temperature in urban cities which is higher than that in suburbs is called Urban Heat Island (UHI), which is put forward by Lake Howard[1] in 1833. This paper chose split window algorithm to retrieve land surface temperature after comparing with single window algorithm, split window algorithm and multi-band algorithm. And this paper improves the split window algorithm of AVHRR, so as to be suitable for MODIS image data. In this paper, land surface temperature is retrieved by split window algorithm based on MODIS images, and the temporal and spatial characteristics of Beijing urban heat island as well as its influencing factors are also analyzed. The result indicates that there is a clear urban heat island in summer, and it is quite close with vegetation index. Besides, the negative correlation relationship between land surface temperature and PM10 is also emphasizely analyzed.Land surface temperature (LST) plays an important role in energy balance of land surface and widely used in quite many fields. For example, soil moisture survey, forest fire monitoring, discrimination of geothermal locations, military camouflage applications and even deposit research, all these can not be separated from the land surface temperature. With the development of remote sensing applications, Single channel algorithm, split window algorithm and multi-band algorithm are the three algorithms of land surface temperature retrieval of existing thermal infrared remote sensed data. At first, this paper chooses split window algorithm after comparing the three algorithms and improves the split window algorithm of AVHRR, so as to be suitable for MODIS image data. This split window algorithm is based on the thermal radiation data which is observed by NOAA—AVHRR. With the basis of thermal radiation function, the brightness temperature can be translated from the two corresponding thermal channels (channel 4 and 5) of AVHRR, and that is to say: Choose MODIS 31(10. 780- 11. 280μm)and 32(11. 770- 12. 270μm)to take the place of the thermal channels (4,10.5-11.3μm and 5,11.5-12.5μm)of AVHRR. Secondly, this thesis discusses how to determine atmospheric transmittance and land surface emissivity, which are the two basic parameter of split window algorithm. Atmospheric transmittance can be retrieved by MODIS 2 and 19 with the relationship of the moisture of atmosphere and atmospheric transmittance. Thirdly, Land surface emissivity mainly depends on the structure of land surface substance, especially the vegetation index, and so land surface emissivity can be obtained by the land surface classification and vegetation index. Finally, land surface temperature can be retrieved through the land surface temperature model, and then the thesis analyses the retrieval precision and error causes.The result indicates that there is a clear urban heat island in summer and autumn, especially summer, with its maximum of not only the intensity but also the area. UHI seems does not exist in spring. In winter, urban cold island effect appears and takes the place of urban heat island. Another conclusion: the land cover types are sensitive to urban heat island effect, and there is a significant negative correlation between vegetation index and distribution of the urban heat island. That is to say: the higher land surface vegetation cover, the less urban heat island is. Therefore, land surface vegetation cover plays an important role on reducing urban heat island.Atmospheric aerosol refers to particles of small-scale range of 0.001-20μm size of liquid or solid caused by natural processes and human activities, which are suspended in the atmosphere. Aerosol plays an important role on weather and climate, while impacts the living environment of mankind. This paper takes Beijing area as an example ,the aerosol optic depth of Beijing has been tried to retrieve using DDV(Dark Dense Vegetation) method from MODIS 1B data. Firstly , the Look Up Table is built by 6S , the dark dense pixel is checked based on NDVI and the retrieved AOD(Aerosol Optical Depth) is validated by the comparation. The results show the DDV method can be used to monitor the urban aerosol very well during the summer in Beijing. analysis of geographic situation combined with Beijing and its surrounding area, the specific is as follows: AOD has higher value in densely populated areas, transportation, and areas of industrial and frequent commercial activities. High-resolution satellite remote sensing results has potential value in monitoring the distribution of emission sources,and can be characterized to the status of air quality in Beijing on a certain extent.Compared the land surface temperature to aerosol optical depth of Beijing on the horizontal scale in the space, this paper taking summer as an example, and find both land surface temperature and AOD have higher value in densely populated areas, transportation, and areas of industrial and frequent commercial activities, but urban heat island and air pollution of Beijing come mainly from the city itself, and temperature effects AOD is not great. the temperature can be characterized urban heat environment on a cetain extent. AOD after revising the vertical and humidity impact can be used to monitor the distribution of particulate matter pollutants in the ground as an effective means,and AOD can be the characterization of urban pollution. Respirable particulate matter is the most important air pollutant of Beijing at present. This paper combines the Municipal Environmental Protection Agency of Beijing air quality conditions and ground measurements, and analyses the relationship between temperature and respirable particulate matter from the microscopic, and finds that the temperature affectes the number of respirable particulate matter little. The number of particles can be decreased as the temperature increases,that is to say: The higher the temperature is, the lower the number of respirable particulate matter is.更多还原