红外与微波遥感一体化反演大气温湿廓线的理论和方法研究

At present, the accuracy and vertical resolution of satellite remotely sensed global atmospheric temperature and humidity profiles cannot meet the needs of the Earth surface parameter retrievals, the numerical prediction models, the atmospheric environmental monitoring applications and so on. This project will carry out the study on the theories and methods of atmospheric temperature and humidity profile retrievals using both infrared and microwave (1—1000GHz) remote sensing. The main research contents of this project are optimal combinations of infrared and microwave bands, microwave sea surface emissivity modeling, microwave land surface emissivity retrievals, developments of the statistical and physical algorithms to retrieve atmospheric temperature and humidity profiles. The objective of this project is to achieve a series of theoretic breakthroughs, leading to the improvements of the retrieval accuracy and vertical resolution of atmospheric temperature and humidity profiles in global scale, which can, respectively, meet or be better than the accuracy and vertical resolution of the radiosonde, i.e., the root mean square errors of the derived temperature profiles are less than 1K for 1km-depth atmospheric layer, and the accuracies of the estimated atmospheric humidity profiles in troposphere reach 10%.

目前，卫星遥感全球大气温度和湿度廓线的精度和垂直分辨率无法满足地球表面参数反演、数值预报模式、大气环境监测等应用的需要。该项目将开展红外与微波（1—1000GHz）遥感一体化反演大气温度和湿度廓线的理论和方法研究，主要研究内容有：一体化反演大气温度和湿度廓线的最佳红外与微波波段组合、微波海表比辐射率模拟、微波地表比辐射率反演、大气温度和湿度廓线反演的统计算法和物理算法发展。该项目的目标是在理论上取得若干突破，使全球大气温度和湿度廓线反演的垂直分辨率与无线电探空的垂直分辨率相当，而反演精度达到或好于无线电探空的精度，即1km气层厚度的温度探测均方根误差小于1K，对流层湿度的探测精度达到10%。

目前，卫星遥感全球大气温度和湿度廓线的精度和垂直分辨率无法满足地球表面参数反演、数值预报、大气环境监测等应用的需要。本项目使用SARTA (Stand-Alone Radiative Transfer Algorithm)快速透过率计算模型和MonoRTM微波大气参数计算程序，结合SeeBor V5大气廓线数据库分别开展红外与微波大气辐射传输模拟实验，根据最优化是信号自由度的函数的评价标准，从理论上确定反演大气温湿廓线的红外(4.5~15.0 μm)和微波波段(50~61 GHz、118.75 GHz和183.31 GHz)。基于Wu和Smith于1997年发展出的红外海表比辐射率模型和FASTEM (FAST microwave sea surface Emissivity Model)模型，对各个影响因素进行敏感性分析，计算得到适合本研究的3.0~15.0 μm光谱区间的红外海表比辐射率和50~61 GHz、118.75 GHz和183.31 GHz微波海表比辐射率。本项目还发展出反演大气温湿廓线的快速局域线性回归算法和非线性物理迭代算法，用所选红外和微波波段联合反演大气温湿廓线的精度分别是~1.0 K和~10%，基本达到项目研究目标。本研究发展出的快速局域线性回归算法应用于经过交叉辐射定标后的“风云三号”B星(FY-3B)红外分光计、微波温度计和微波湿度计数据，红外与微波观测联合反演全球大气温湿度廓线，并用实测RAOB (The Universal RAwinsonde OBservation program)探空数据、ECMWF (European Centre for Medium-range Weather Forecasts)再分析数据和AIRX2RET V6产品对反演结果进行验证，取得了较好的反演精度。与MOD07_L2 V5大气廓线产品相比，FY-3B红外与微波探空仪数据联合反演结果的精度更高：在800 hPa以下的近地表层，大气温度廓线反演误差减小了1.5 K，而大气湿度廓线反演误差减小了0.5 g/kg。