基于环境卫星数据的南京地表温度反演及时空格局分析

发布时间：2018-08-28 17:08

【摘要】：地表温度(Land Surface Temperature,LST)是研究全球气候变化的关键参数,是地表能量交换、地球中各种循环过程中的主导因子,与人们的生活息息相关。随着全球城市化进程的不断加快,环境问题日益凸显,城市热岛效应已经成为一个重要的研究内容。本文利用遥感技术,在ENVI、ArcGIS等软件的支持下,以快速发展的江苏省省会城市南京作为研究区域,基于HJ-1B遥感卫星影像,采用辐射传输法(RTE)、覃志豪单窗算法(Qins')和普适性单通道算法(JMS)进行地表温度的反演。对比反演的结果,在选取精度较高的反演方法的基础上分析南京市地表温度的空间分布特征,并分析地表温度与土地利用情况和景观指数的关系。通过2010年夏季、2013年四季五个时相的地表温度分布图,动态分析了两个年份南京市地表温度的年际变化和季节变化,并在此基础上分析了土地利用情况及植被指数等对地表温度的影响。得到的主要结论如下:(1)三种算法均能较好反应南京地区的地表温度趋势,市中心城区地表温度最高,向四周郊区及植被覆盖地区递减。但三种算法得到的地表温度结果仍存在一定差异,RTE得到的温度反演结果最高;Qins'反演结果与其较接近,温差在2K左右;JMS反演结果相对较低,与RTE反演结果相差10K左右。而与MODIS温度产品相比,RTE算法精度最高,与MODIS地温产品的温度差值多集中在2.1K;Qins'的反演结果略低,温差多集中在3.87K;而JMS的结果明显偏低,温差多集中在5.96K。(2)空间上,南京市的地表温度由城区中心向郊区逐渐递减。主城区的高温区与较高温区占了 60%以上的面积;北部六合区、西部浦口区、南部高淳区温度等级分布类似,以较低温和中温为主。靠近主城区的江宁区和溧水区,其温度等级多为中温区和较高温区。在研究地温空间分布的基础上,选择影响因子,探讨其与地表温度的关系,发现:地表温度与耕地覆盖比例的相关性最大,其次为板块密度(PD),接下来依次是建设用地面积比例(PerBuild),板块面积(AREA),最大板块指数(LPI)。而植被指数中的植被覆盖度(Pv)和归一化植被指数(NDVI)与地表温度关系密切,Pv与LST的回归系数R2=0.64,NDVI与LST的回归系数R2=0.51。相对于NDVI,Pv能更好的预测因植被覆盖影响的地表温度。(3)时间上,自2010年至2013年,主城区高温区、较高温区面积明显增加,而较低温区、中温区面积减少;近郊区的江宁区,低温区和较低温区有所减少,而较高温区和高温区有所增加;浦口区与江宁区变化规律类似,低温区和较低温区减少,较高温区与高温区增加。整体体现由中间向两极转化,温度等级差异性增加,城市热岛效应更加显著。从本市不同的土地利用情况来看,随着时间的变化,南京市2010-2013年间,研究区内的自然景观减少,人工的景观有所增加。随着城市的不断扩张,热岛范围会不断增大,土地利用状况的改变会对地表温度的空间分布造成很大影响,其中城市建设用地贡献最大,是形成热岛的主要原因,而水体和林地有比较好的降温作用。在探讨植被指数与地表温度的关系时发现,在空间分布上LST与NDVI、Pv变化趋势几乎是相反的(水体除外)。从季相变化特征上看,南京市春季和夏季的热岛效应明显,特别是夏季,而其他两个季节热岛效应并不明显。随季节的变化,不同的植被覆盖情况对地表温度等的影响也是有所不同的。LST在春、夏、秋三季与NDVI、Pv的分布趋势相反,植被覆盖度高的地区,地表温度相对较低,而冬季LST与NDVI、Pv的值都普遍偏低。
[Abstract]:Land Surface Temperature (LST) is a key parameter in the study of global climate change. It is also a leading factor in the process of energy exchange on the earth's surface. It is closely related to people's life. Based on the HJ-1B remote sensing satellite image, this paper uses remote sensing technology, ENVI, ArcGIS and other software support to retrieve the surface temperature of Nanjing, the capital city of Jiangsu Province, using the radiation transfer method (RTE), the Qins'algorithm and the universal single channel algorithm (JMS). The results show that the spatial distribution of land surface temperature in Nanjing is analyzed on the basis of the inversion method with high precision, and the relationship between land use and land use and landscape index is analyzed. The main conclusions are as follows: (1) The three algorithms can better reflect the trend of surface temperature in Nanjing. The surface temperature in the downtown area is the highest, and decreases to the surrounding suburbs and vegetation coverage areas. There are still some differences in the surface temperature results obtained by RTE, and the temperature inversion results obtained by RTE are the highest; the Qins'inversion results are close to them, and the temperature difference is about 2K; the JMS inversion results are relatively low, which is about 10K different from the RTE inversion results. 1 K; Qins'inversion result is slightly lower, the temperature difference is mostly concentrated in 3.87K; while JMS result is obviously lower, the temperature difference is mostly concentrated in 5.96K. (2) The surface temperature of Nanjing decreases gradually from the center of the city to the suburbs. On the basis of studying the spatial distribution of geotemperature, the influencing factors were selected to study the relationship between the spatial distribution of geotemperature and the surface temperature. It was found that the correlation between the surface temperature and the proportion of cultivated land coverage was the greatest, followed by the plate. The vegetation coverage (Pv) and normalized vegetation index (NDVI) in the vegetation index were closely related to the surface temperature. The regression coefficients of Pv and LST were R2 = 0.64, and that of NDVI and LST was R2 = 0.51. It is predicted that the surface temperature affected by vegetation cover will increase from 2010 to 2013 in the high temperature area of the main urban area, while decrease in the middle temperature area in the low temperature area. Similarly, low-temperature and low-temperature zones decrease, while high-temperature and high-temperature zones increase. The overall embodiment is from the middle to the two poles, the difference of temperature grades increases, and the urban heat island effect is more significant. With the continuous expansion of the city, the scope of the heat island will continue to increase. The change of land use status will have a great impact on the spatial distribution of the surface temperature. The urban construction land contributes the most to the formation of the heat island, and the water and woodland have a better cooling effect. It is found that the variation trend of LST, NDVI and Pv is almost opposite in spatial distribution (except water body). The heat island effect is obvious in spring and summer, especially in summer, but not in the other two seasons. The distribution trend of LST is opposite to NDVI and Pv in spring, summer and autumn. The surface temperature is relatively low in areas with high vegetation coverage, while the values of LST, NDVI and Pv are generally low in winter.
【学位授予单位】：南京农业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P412.27

【相似文献】