Tang Xinming, Yao Jiaqi, Li Guoyuan, Ai Bo, Wang Zhenming, Gao Xiaoming
Appl Opt. 2020 May 1;59(13):4064-4075. doi: 10.1364/AO.385808.
Satellite laser altimetry can obtain submeter or even centimeter-level surface elevation information over a large range. However, the laser will inevitably be affected by clouds during transmission through the atmosphere, which seriously affects the accuracy of altimetry. In this paper, based on laser altimetry data, cloud optical depth inversion was realized by using the Fernald method. The influence of clouds on the echo waveform data was analyzed with actual data, and a method of cloud scattering error correction was proposed. The existing error correction methods are mostly based on the results of semi-analytical Monte Carlo simulations. In observations, it is difficult to synchronously obtain the parameters required for simulation, which significantly limits the method. Therefore, a method for correcting the cloud scattering error of satellite laser altimetry data based on an exponential model is also proposed. The experimental results show that when the cloud optical depth is 0-2, the root mean square error of the model is 0.05, which can correct the height measurement deviation caused by the cloud to within 5 cm and improve the availability of the laser height measurement data affected by the cloud scattering.
卫星激光测高能够在大范围获取亚米甚至厘米级的地表高程信息。然而，激光在大气中传输时不可避免地会受到云层影响，这严重影响测高的精度。本文基于激光测高数据，利用费尔纳德方法实现了云光学厚度反演。结合实际数据分析了云层对回波波形数据的影响，并提出了一种云散射误差校正方法。现有的误差校正方法大多基于半解析蒙特卡罗模拟结果。在观测中，难以同步获取模拟所需的参数，这极大地限制了该方法。因此，还提出了一种基于指数模型的卫星激光测高数据云散射误差校正方法。实验结果表明，当云光学厚度为0至2时，模型的均方根误差为0.05，可将云层引起的测高偏差校正到5厘米以内，提高受云散射影响的激光测高数据的可用性。
