研制了一套船载海洋激光雷达，用于探测海水光学参数垂直廓线。2017年8月，该系统在黄海海域进行了实验测量。在准单次散射模型中引入原位测量的光学参数，实现了理想激光雷达回波信号的模拟，并将该理想信号与系统响应函数卷积后精确复现了实验的激光雷达信号。采用Fernald后向迭代积分法（简称Fernald法），比较了不同水体悬浮物激光雷达比下反演的激光雷达衰减系数α与原位漫射衰减系数K_d的差别。基于停航时标定的水体悬浮物激光雷达比，采用Fernald法获得了走航时的激光雷达衰减系数。进一步地，提出一种基于米散射激光雷达数据和原位测量的后向散射数据的融合算法，模拟了高光谱分辨率激光雷达（HSRL）反演α的过程，并将其与Fernald法进行了比较。实验结果表明，自研的海洋激光雷达能够有效探测海水光学参数，基于合适的水体悬浮物激光雷达比的Fernald法可以有效应用于米散射激光雷达的反演，未来无需假设的HSRL在海水光学参数探测领域具有更大的优势。

Studying seawater optical properties is of great importance in global climate change and material cycle. lidar has the ability to retrieve the profiles of seawater's optical properties. A single-scattering lidar equation is typically a useful and effective model of lidar return. However, lidar return depends on the volume scattering function at 180° scattering angle and lidar attenuation coefficient, which makes retrieval from an equation difficult. The Fernald method is often used to retrieve backscatter lidar return with the assumption of lidar ratio. High Spectral Resolution Lidar (HSRL) can retrieve optical properties without assumption.
A shipborne traditional lidar was developed to detect the vertical profile of seawater optical parameters. Experiments on seawater were conducted in the nearshore and offshore regions of the Yellow Sea. The lidar system was fixed on the front deck of a scientific survey boat. In situ optical measurements were also performed in the two regions. A simple quasi-single-scattering approximation was employed to calculate a modeled lidar return with inherent optical properties derived from the in situ measurement. The comparison of oceanographic lidar returns with modeled lidar returns using nearly coincident in situ optical properties were in perfect agreement with the nearshore and offshore regions, indicating that lidar can effectively detect seawater optical parameters.
The difference between the inverse lidar attenuation coefficient and the in situ diffusion attenuation coefficient were analyzed based on the Fernald method with different lidar ratios. With the use of the calibrated lidar ratio, the lidar attenuation coefficients while sailing were obtained by using the Fernald method. A fusion algorithm based on traditional lidar data and in situ backscatter coefficient was also proposed to simulate HSRL. Then the accuracies of the Fernald method and fusion algorithm were compared. In the nearshore water column, diffuse attenuation coefficient varied from 0.15 m⁻¹ to 0.28 m⁻¹, and the maximum error of both methods was below 11%. As for the offshore water column, diffuse attenuation coefficients changed little through depths, approximately 0.1 m⁻¹ to 0.16 m⁻¹. The maximum error of the two methods was nearly 17%. The statistical analysis showed that diffuse attenuation coefficient can be well employed both by fusion algorithm and the Fernald method (with calibrated lidar ratio).
This paper described the applications of lidar for profiling the properties of upper ocean. To overcome the assumption of lidar ratio in the future while retrieving water column information from traditional lidar, the HSRL without assumption has a great advantage in the field of seawater optical parameter detection.