主被动观测系统下云和气溶胶辐射偏振态逐阶演变机理研究

With rapid developments in polarization measurement instruments, the correct interpretation of three-dimensional radiation polarization effects has become the key link in the development of the remote sensing inversion technologies of cloud and aerosol parameters in the active and passive polarization observation systems. However, the observed signals, involving three-dimensional polarization radiation transport, give various complicated three-dimensional polarization radiation phenomenon, which one-dimensional radiative transfer cannot take into account. To this end, the project intends to study cloud and aerosol successive-order scattering radiation polarization state evolution mechanism using numerical simulation and theoretical analysis. Firstly, the observation simulation modules of space- and ground-based polarization lidar and passive polarimeter are added to three-dimensional cloud and aerosol polarized radiative transfer model MSCART. The variance reduction method is developed to improve and optimize the numerical performance of the Monte Carlo forward and backward simulation. Then, numerical simulation study of cloud and aerosol successive-order scattering radiation polarization state is carried out to reveal the formation mechanism of three-dimensional polarization effects by multiple scattering and horizontal transport in clouds and aerosol atmospheres, so as to interpret the various observing three-dimensional polarization phenomena. The purpose of this study is to lay a theoretical foundation for the reliability assessment of the international operational cloud and aerosol polarization remote sensing algorithms and the future high spatial and temporal resolution retrieval of cloud and aerosol properties using three-dimensional polarized radiative transfer theories.

随着偏振探测仪器快速发展，正确解译三维偏振辐射效应成为主被动偏振观测系统下云和气溶胶参数遥感反演技术发展提高的关键环节，但是观测信号涉及三维偏振辐射传输，形成了各种纷繁复杂、平面平行辐射传输理论难以解释的三维偏振辐射现象。为此，本项目拟采用数值模拟和理论分析方法，开展云和气溶胶辐射偏振态逐阶演变机理的研究。首先在三维云大气辐射传输模式MSCART基础上添加空基和地基偏振激光雷达和被动偏振辐射仪的观测模拟模块，开发方差缩减方法完善和优化蒙特卡罗前向和后向数值模拟性能，然后由此展开云和气溶胶逐阶散射辐射偏振状态贡献的数值模拟研究，揭示主被动观测系统下三维偏振辐射效应的云和气溶胶多次散射和水平输运的形成机理，达到解译不同三维偏振辐射现象的目的，为国际业务云和气溶胶偏振遥感算法可靠性评估和未来三维偏振辐射传输反演云和气溶胶高时空分辨率特性打下理论基础。

正确解译天基和地基主被动偏振观测数据是优化当前和发展未来云和气溶胶大气遥感探测技术的关键环节之一。虽然当前已有相关学者基于三维偏振辐射传输模拟开展了偏振辐射信号的气溶胶和云参数敏感性研究，但是在偏振辐射信号背后，气溶胶和云多次散射过程产生的偏振状态逐阶演变的机理则鲜有研究，亟待开展。针对于此，本项目首先构建了耦合先进逐阶合成方差缩减方法的通用前向和后向蒙特卡罗偏振辐射传输理论框架，提出了斯托克斯矢量和穆勒矩阵权重跟踪方案，实现了开源三维云大气偏振辐射传输模式MSCART的天基和地基偏振激光雷达和被动偏振辐射仪的观测模拟功能。在此基础上，本项目开展了大量的云和气溶胶逐阶散射偏振辐射信号贡献的数值模拟实验研究，揭示了解译偏振辐射信号的关键在于确定哪种散射路径在偏振辐射信号中占主导以及在此种路径中散射平面的旋转角和散射角如何分布，由此创新性地提出了双平面多次小角多次散射路径模型和相对主散射平面的单侧弯曲和双侧随机扭动多次散射路径模型，依据斯托克斯-穆勒矩阵的数学描述，推导了观测穆勒矩阵与各阶散射相矩阵和旋转矩阵之间的数学表达式，重点阐述了多次散射路径的几何特性（特别是旋转角的路径分布）在改变辐射信号偏振状态、增强和减弱线和圆偏振度上的重要性，由此成功解释了多个不同观测条件下穆勒矩阵、斯托克斯矢量、线偏振度和圆偏振度等偏振辐射信号的时空分布特征，例如在不同水平或垂直辐射传输距离情况下激光雷达液态云层后向散射穆勒矩阵多重叶状结构和径向分布特征、被动辐射仪观测到的云边斯托克斯矢量U分量跃变结构分布特征。本项目提出的多次散射路径模型可以为偏振辐射传输理论反演高时空分辨率云和气溶胶特性提供理论基础。