基于Landsat8数据的地表温度反演算法研究——以乌鲁木齐市为例

Research on Surface Temperature Inversion Algorithm Based on Landsat8 Data in Urumqi City

不同的地面温度反演算法会影响地面温度的反演精度。根据2016年7月Landsat8遥感影像,以乌鲁木齐市为研究区。采用大气校正法和单窗算法反演研究区域的地表温度,同时与亮温进行对比。结果表明:大气校正法和单窗算法的陆地表面温度和亮温反演在地理上的分布很接近,但也有不同。两个算法之间的平均温差为0.35℃,单窗算法比平均亮度温度低0.05℃,大气校正法比平均亮度温度低0.4℃[1]。大气校正法较单窗算法精度低。

Different surface temperature inversion algorithms affect the precision of inverting the surface temperature.Based on the Landsat8 image in July 2016,taking Urumqi as the research object,the surface temperature of the study area is retrieved using the the momo-window algorithm and the radioactive transfer equation algorithm which are compared with its brightness temperature.The consequence is surface temperature's geographical distribution detected by the momo-window algorithm and the radioactive equation algorithm is approximately the same as the brightness temperature,but there are also differences.The average temperature difference between the two algorithms is 0.35℃.Compared with the brightness temperature mean value,momo-window algorithm is 0.05℃lower than the radioactive transfer equation algorithm.The atmospheric correction method is 0.4℃lower than the average brightness temperature[1].The accuracy of the momo-window algorithm is higher than that of the radioactive transfer equation algorithm.