高光谱分辨率激光雷达的气溶胶多参数反演及其辐射估算应用研究

High-spectral-resolution Lidar (HSRL) can obtain the high-precision and high-resolution atmospheric profile effectively, which has been listed as the payload of the second-generation space-borne Lidar by several countries. Compared with the development of hardware technology, the data processing methods of HSRL still need to be improved. Moreover, the application of data products remains to be tapped explored. In this proposal, the non-linear relationship between the extinction coefficient and the backscattering coefficient was will be used to construct a regularized atmospheric multivariate parameters inversion system and propose develop the corresponding improved methods. To improve the regularized operation speed and inversion accuracy of regularization, the long-term ground-based observation data is used as prior knowledge, and the correlation between multivariate atmospheric profile and atmospheric parameters is used as posterity knowledge. The project will also use the data products of the HSRL : (atmospheric vertical distribution information, optical and physical parameters) . These parameters were used as input for the radiative transfer model to study the aerosol radiative on effects under different atmospheric environmental quality in the typical cities over Central China. In this project, the detection advantages of the HSRL will provide some theoretical references for the development of the algorithm for the second-generation satellite-borne lidar in China. At the same time, it will also better further reveal the mechanism of the influence of aerosol optical and physical characteristics on the short-wave and its sub-band radiation radiative effects.

高光谱分辨率激光雷达能有效获取高精度、高分辨率的大气廓线探测数据，已被多国列为二代星载大气探测激光雷达的有效载荷。与硬件技术的发展相对比，其数据处理方法有待提升、数据产品应用有待挖掘。本项目充分利用高光谱分辨率激光雷达消光系数廓线与后向散射系数廓线具有非线性关系的特点，构建正则化多元大气参数反演系统，提出相应的改进方法，并利用地面台站长时序观测数据作为先验知识、多元大气廓线与大气参数相关性作为后验知识，实现正则化运算速度和反演精度的提高。项目还将利用高光谱分辨率激光雷达的数据产品：大气垂直分布信息，光学和物理参数作为辐射传输模式的输入，开展不同大气环境质量下，华中地区城市型气溶胶辐射效应研究。本项目的实施，将充分发挥高光谱分辨率激光雷达的探测优势，可为我国星载激光雷达载荷的算法开发提供部分理论参考，同时，还将更好的揭示华中地区气溶胶光学、物理特性变化对短波及其各子波段辐射效应的作用机制。

随着社会经济发展，我国大气遥感探测目标在逐渐发生改变，十二五、十三五期间，大气中细粒子颗粒物的减排是大气环境治理核心任务，激光雷达的廓线观测发挥着重要作用，它能有效填补星载被动影像缺失的污染物垂直分布信息，为污染成因、传输扩散和预报预测提供重要的参数。尽管十四五，国家对大气遥感探测目标做出了调整，例如PM2.5与O3的相关性研究、“双碳”目标涉及的温室气体监测等，但实际上气溶胶粒子的多参数反演，仍然是上述大气遥感，乃至对地定量遥感的基础。本项目紧扣气溶胶多参数反演的目标开展相关研究，在气溶胶反演算法中引入机器学习，提高激光雷达有限波段的信息利用价值；利用台站长时期地面观测数据做约束，为大气参数反演提供必要的先验知识。此外，项目组利用高精度气溶胶参数作为模型输入，深化气溶胶辐射效应的研究；深入分析气溶胶对大气边界层高度变化的影响作用，及其对污染物传输的反馈和抑制。项目的执行揭示了区域气溶胶光学特性的变化规律、分析与京津冀-珠三角-长三角显著不同的气溶胶辐射作用、提出了基于粒子特征的气溶胶边界层查找方法。不仅如此，本项目还支撑团队开展系列国产卫星的反演工作（包括：风云3D和高分5号），在开展理论创新的同时，落实了理论方法的工程应用价值，为国产卫星气溶胶多参数反演提供了具体的解决方案，其成果可为精确大气校正、地表生态参数反演，铺垫理论基础。