中国农业气象 ›› 2011, Vol. 32 ›› Issue (增刊): 148-152.

• 论文 • 上一篇    下一篇

基于MODTRAN 4的贵阳市Landsat热红外波段大气透过率估计

谷晓平,孟维亮,于飞,王备,敖芹   

  1. 中国气象局成都高原气象研究所开放实验室,成都610071;贵州省山地气候与资源重点实验室,贵阳550002;
    贵州省山地环境气候研究所,贵阳550002)
  • 出版日期:2011-10-31 发布日期:2011-11-24
  • 基金资助:

    贵州省优秀科技教育人才省长专项资金项目(黔省专合字\[2008\]70号);中国气象局成都高原气象开放实验室基金项目(LPM2009007);黔科合人才团队项目(20104012)

Estimation of Atmospheric Transmittance in Thermal Infrared in
Guiyang Band from Landsat on MODTRAN 4

GU Xiaoping, MENG Weiliang, YU Fei,WANG Bei, AO Qin   

  1. Open Laboratory, Institute of Plateau Meteorology, CMA, Chengdu, Chengdu 610071,China; Guizhou Key Laboratory of Mountainous〖JP〗
    Climate and Resource, Guiyang550002; Guizhou Institute of Mountainous Environment and Climate, Guiyang550002)
  • Online:2011-10-31 Published:2011-11-24

摘要: 大气水汽含量对热红外波段大气透过率的干扰是遥感影像反演地面温度的主要影响因素之一,贵阳市夏半年气温低、水汽含量高的气候特点增加了大气水汽的干扰作用。为实现Landsat卫星资料在贵阳市地面温度反演中的应用,完善大气校正,提高反演精度,本文通过MODTRAN 4模拟了22℃时Landsat ETM+影像热红外波段的大气透过率随水汽含量的变化规律,拟合了利用水汽含量计算该波段大气透过率的分段式计算方程。经模拟演算和地温反演应用,结果表明:估计方程的演算误差在0.9~1.3℃,满足贵阳市地表温度反演的误差要求;实际应用中贵阳市地温反演结果与台站资料计算的地温值存在2.7℃的误差,还存在其它参数误差,反演精度有待进一步提高。

关键词: 地表温度反演, 大气模型, 大气水汽含量, 大气透过率

Abstract: The interference on the atmospheric transmittance of thermal infrared band, generated by atmospheric water vapor content, is one of the main factors for retrieving land surface temperature from remote sensing images. In Guiyang, the climatic characteristics of low temperatures and high water vapor content in the summer half year increases the interference. To retrieve land surface temperature in Guiyang from Landsat satellite data and improve retrieval accuracy, in this paper, we simulated the variation of atmospheric transmittance with atmospheric water vapor content in Landsat ETM+ thermal infrared band at 22℃ on MODTRAN 4, and fitted a set of piecewise equations for calculating atmospheric transmittance in the band from atmospheric water vapor content. After simulation algorithms and applications, the results showed that calculation error of the piecewise equations was about 0.9~1.3℃, which could meet the error requirement of retrieval accuracy in Guiyang. In the practical application of retrieving land surface temperature in Guiyang, there was about 2.7℃ error between retrieval results and temperature calculated on weather station data, we needed to find out other errors and further improve the retrieval accuracy.

Key words: Land surface temperature retrieval, Atmospheric model, Atmospheric water vapor content, Atmospheric transmittance