Abstract: Retrieving atmospheric water vapor from remote sensing data is very important in such aspects as water circulation, global climatic change, weather forecast and atmosphere correction. Considering the scanning angle of the sensor, the two-band ratio method for retrieval of atmospheric total water vapor is developed by moderate-resolution imaging spectro-radiometer(MODIS) near infrared measurements. The RMS error compared to radiosonde observation is 0.29 g/cm2, and the correlation coefficient is 92%. The comparison of the monthly mean total water vapor distribution shows that the retrieval water vapor field is in good agreement with that of the MODIS product.

Key words: MODIS, MODTRAN, total precipitable water

摘要： 遥感反演大气水汽总含量对于研究水循环、全球气候变化、天气预报、遥感大气校正等具有重要意义。本文考虑了卫星扫描视场角的影响，实现并改进了用MODIS近红外资料反演大气水汽总含量的两通道比值反演模式，并用辐射传输模式定量讨论了误差。结果表明，反演值和探空均方根误差RMS=0.29 g/cm2，相关系数在92%。对月平均水汽分布的比较显示，反演水汽总含量与MODIS产品的分布是一致的，且更为合理。

关键词: MODIS, MODTRAN, 水汽总含量