HY-2A卫星散射计降雨影响及其校正模型研究

The scatterometer onboard HY-2A satellite is the first operational spaceborne microwave scatterometer in China, which can retrieve and monitor the global ocean surface vector winds.The related validations demonstrated that the wind retrieval accuracy of the HY-2A scatterometer meets its design requirement in case of no rain. However, in raining conditions, the wind retrieval errors become large significantly.The comprehensive impact of rain including attenuation,volume backscattering,and surface perturbation induces a dramatic quality degradation of the retrieved wind fields in raining area.In most cases,the degraded wind data can not fulfill the precision need of the operational applications. Aiming at this problem, this project intends to use the radar backscattering coefficients from HY-2A scatterometer,rain rate data from TRMM(Tropical Rainfall Measuring Mission)/PR(Precipitation Radar),ocean surface wind speeds from the SSM/I radiometer,and ocean surface wind directions from ECMWF(European Center for Medium-Range Weather Forecasting) to constuct a collocated data set.Based on some of the collocated data set, the impacts of rain on the radar backscatter coefficients are to be analyzed ,and a wind-rain coupled scattering model and the corresponding rain correction model are derived by the method of radiation transfer equation.On the basis of modeling, a rain-induced noise level model for the HY-2A scatterometer in case of precipitation are proposed. Furthermore, a wind retrieval algorithm is established for the HY-2A scatterometer in order to accurately retrieve vector winds in areas of rain.Using the rain correction model and the new wind retrieval algorithm, the rain effect of the backscatter coefficients can be quantitatively adjusted in process of wind retrieval, and the wind retrieval accuracy of the scatterometers from the HY-2A satellite and its successors(HY-2B/C) would be improved in areas of rain.

HY-2A卫星散射计是我国第一个业务化运行的星载微波散射计，可实现全球海面风场的反演监测。相关检验表明，在无雨条件下，HY-2A散射计的风场反演精度能够达到其设计要求，而在降雨条件下反演误差显著增大。降雨衰减、雨滴的体散射、以及雨滴对海表面的冲击扰动等综合影响，使得降雨区域风速和风向反演质量急剧下降，不能满足业务化应用的要求。针对此问题，本项目拟利用HY-2A散射计获得的后向散射系数实测数据和TRMM/PR测得的降雨率、SSM/I辐射计测得的海面风速、以及ECMWF数值模式风向等数据，通过时空匹配构建建模和验证数据集，研究降雨对后向散射系数的影响，并构建风雨联合散射模型及降雨校正模型。在此基础上，通过构建散射计降雨噪声测量误差模型，建立降雨条件下的海面风场反演算法，实现风场反演过程中后向散射系数降雨影响的定量化校正，改善HY-2A及其后续卫星(HY-2B/C)散射计在降雨区域的风场反演。

微波散射计是目前全球海面风场探测的主要星载传感器。实验表明，在无雨条件下，其风场反演精度能够达到设计要求，而在降雨条件下其风场反演质量急剧下降，不能满足业务化应用的要求，从而导致降雨区域有效风场数据的缺失。. 本项目以我国HY-2卫星为研究对象，利用其微波散射计测得的后向散射系数数据、欧洲中期天气预报(ECMWF)数值模式风速和风向数据、以及SSM/I辐射计测得的降雨率数据，通过时空匹配构建了同步建模数据集和验证数据集，分析了降雨率对散射计风速反演误差和后向散射系数的影响规律，并在此基础上利用人工神经网络建立了散射计风速降雨校正模型和后向散射系数降雨校正模型，以及考虑降雨因子的海面风场反演算法，探讨了降雨对散射计测风的影响机理，形成了初步的理论模型框架。此外，本项目针对HY-2卫星散射计的特点，利用KNN(K Near Neighbor)算法构建了一套适用于HY-2卫星散射计的降雨识别模型（HY2RRM），同时利用BP(Back Propagation)神经网络构建了散射计风速和降雨率联合反演模型，为基于散射计数据的风速和降雨率的同步反演进行了有益的探索和尝试。. 实验结果表明，本项目所构建的风速校正模型和后向散射系数校正模型，在中等风速条件下，能够显著减小降雨所引起的反演风速和后向散射系数的系统性偏差，有效消除了降雨条件下散射计后向散射系数测量值和反演风速偏高的问题。KNN降雨识别算法在不依赖于外部辅助信息的条件下，获得了较高的降雨识别准确率，为HY-2卫星散射计观测数据的降雨识别提供了一种有效的技术手段。基于神经网络的风速和降雨率联合反演模型，能够实现风速和降雨率的同步反演，同时对去除降雨影响造成的风速反演偏差具有明显的效果。. 本项目构建的风速降雨校正模型、后向散射系数降雨校正模型、降雨自动识别模型、以及基于神经网络的风速和降雨率联合反演模型等成果可用于我国HY-2系列卫星微波散射计观测数据业务化风场反演中的降雨标识及降雨影响校正，提高降雨条件下的海面风场反演精度，拓展降雨区域的风场反演能力。