基于扇形波束扫描微波散射计的高分辨率风场反演研究

Currently, the operational scatterometers generally produce mesoscale sea surface winds with a true spatial resolution of ≥ 25 km. Therefore, scatterometer winds are not effectively explored in some high-resolution applications, such as coastal circulation modeling and meso-/small-scale dynamics research. A new design of scatterometer, namely rotating fan-beam scatterometer (RFSCAT), is now under developing in China. RFSCAT combines the observation characteristics of the traditional two types of scatterometers, i.e., the fixed fan-beam scatterometers and the conically scanning pencil-beam scatterometers, such that it produces a larger amount of high-resolution (in particular, the range resolution is better than 5 km) backscatter measurements with diverse azimuth and incidence angles for each wind vector cell. This challenges the current wind inversion procedures, but also presents opportunity for producing higher quality wind products with an improved quality control. This project investigates the spatial responses of RFSCAT backscatters at different incidence and azimuth angles, and proposes a new method to aggregate the multi-incidence/azimuth measurements into the high resolution wind grids (grid size < 10 km). Then it presents a sophisticated model to analyze the relationship between the spatial resolution and the measurement accuracy, leading to an optimal trade-off (resolution and accuracy) for the development of RFSCAT high-resolution winds. The main objective of this project is to develop a thorough inversion process for defining high-resolution RFSCAT winds, including unbiased inversion method, rain identification and ambiguity removal algorithms. Finally, such process is used to produce high-resolution sea surface and coastal winds for the upcoming Chinese RFSCATs. The results of this project will enrich the wind observations of different scales, improve our understanding of meso-/small-scale sea surface dynamics, and promote the wind-related applications and scientific researches over the coastal areas.

当前业务化微波散射计的海面风场的真实分辨率通常大于25 km，无法满足近海岸区域和中小尺度海面动力学等研究的需求。我国新型的扇形波束扫描微波散射计（RFSCAT）结合传统的固定扇形波束和扫描笔形波束两种散射计的特点，可以为各个风单元获取更多高分辨率（尤其是距离分辨率优于5 km）的后向散射系数测量，且这组测量具有多方位角和多入射角的特点。本项目基于RFSCAT研究不同入射角和方位角下雷达后向散射的空间响应特征，探讨多入射/方位角测量组合到高分辨率风单元(分辨率优于10 km)的新方式，并权衡分析分辨率和测量精度的关系。重点开发高分辨率风场的无偏差反演、降雨识别和风向去模糊算法，产生相应的RFSCAT高分辨风场（包括近海岸风场）。项目的研究成果有助于丰富不同尺度风场的观测信息，改善人们对中小尺度海面动态过程的认识，促进风场在近海岸区域的科学研究和应用。

星载微波散射计是目前获取全球海面风场最主要的传感器。散射计的风场数据在海洋气象灾害监测、数值天气预报模式以及海气相互作用科学研究中得到广泛且深入的应用。然而，过去和当前卫星散射计业务化海面风场的典型分辨率是25-50 km，无法满足近海岸区域、中小尺度海面动力过程等一些较高分辨率应用的需求。中法海洋卫星散射计（CSCAT）使用扇形波束旋转扫描体制，具有空间分辨率较高、以多角度测量同一海面雷达后向散射系数的特点，为近海岸海面风场反演提供了新的机遇。本项目针对CSCAT近海岸海面风场处理的主要流程和关键技术开展研究。首先，开发了星载扇形波束扫描散射计仿真程序，系统分析了新体制散射计海面后向散射的误差特性及主要的误差源，并利用海洋定标方法得到了垂直（VV）极化和水平（HH）极化的定标系数。发现VV极化的定标系数小于HH极化。其次，对CSCAT风场反演残差特性进行数值分析，利用匹配的降雨数据并对风场质量敏感因子（即反演残差）其进行分区研究，对不同降雨条件下质量敏感因子分布和风场质量的特征进行了详细分析，进而提出了一种CFOSAT微波散射计海面风场质量控制（QC）的方法。质量控制后的海面风场与参考风场具有良好的一致性。最后，利用一种矩形窗算术平均的方法实现近海岸风场反演的快速预处理，有效地实现CSCAT近海岸海面风场反演。通过对比CSCAT、欧洲先进散射计（ASCAT）以及美国QuikSCAT散射计近海岸风场的分布特征和统计特征，发现CSCAT风场的质量在离岸40 km以外区域具有良好的一致性，而在离岸40 km以内显著恶化。分析表明，近海岸区域的风场处理可能仍存在陆地污染的影响，但与模式背景风场和浮标风场相比，总体具有良好的一致性。项目开发的高分辨率的近海岸风场数据可以为近岸船只走航、资源开发、生态和侵蚀过程监测等方面提供重要的数据支撑。