周智标, 周辉, 马跃, 宋越, 李松. ICESat-2激光雷达海面信号提取和海浪要素计算[J]. 红外与激光工程, 2023, 52(2): 20220366. DOI: 10.3788/IRLA20220366
引用本文: 周智标, 周辉, 马跃, 宋越, 李松. ICESat-2激光雷达海面信号提取和海浪要素计算[J]. 红外与激光工程, 2023, 52(2): 20220366. DOI: 10.3788/IRLA20220366
Zhou Zhibiao, Zhou Hui, Ma Yue, Song Yue, Li Song. ICESat-2 lidar sea surface signal extraction and ocean wave element calculation[J]. Infrared and Laser Engineering, 2023, 52(2): 20220366. DOI: 10.3788/IRLA20220366
Citation: Zhou Zhibiao, Zhou Hui, Ma Yue, Song Yue, Li Song. ICESat-2 lidar sea surface signal extraction and ocean wave element calculation[J]. Infrared and Laser Engineering, 2023, 52(2): 20220366. DOI: 10.3788/IRLA20220366

ICESat-2激光雷达海面信号提取和海浪要素计算

ICESat-2 lidar sea surface signal extraction and ocean wave element calculation

  • 摘要: 美国NASA于2018年发射的ICESat-2 (The Ice, Cloud, and land Elevation Satellite-2)卫星上搭载的ATLAS (Advanced Topographic Laser Altimeter System)是目前为止全球唯一一个对地观测的星载光子计数激光雷达,具有较高的轨向空间采样率,为用遥感的方法探测海浪要素提供了可能。光子计数激光雷达用于海浪探测的前提是能够准确地提取来自海面的信号光子,并确定瞬时的海面廓线。迄今为止,用星载光子计数激光雷达探测海面形态和海浪要素的研究鲜见报道,也缺少专门针对海面信号光子的提取方法。基于海面信号光子的分布特点,文中提出了一种新的信号提取算法:首先通过直方图统计及自适应的阈值选取完成对海面回波光子的粗去噪;然后基于激光雷达光斑尺寸和海面波动特点,选取合适的搜索邻域计算信号点和噪声点密度,根据两者点密度差异对信号光子和噪声光子分类;最后用高斯函数拟合的方法进一步去除密度较大的后向散射噪声光子,最终得到来自海面反射的信号光子。利用上述算法提取了太平洋7个不同海况区域的海面信号光子和瞬时海面廓线并进一步计算出当地海浪的峰值波长和周期。将计算结果与同期欧洲中期天气预报中心(European Centre for Medium-Range Weather Forecasts, ECMWF)的全球大气再分析ERA5(ECMWF Re-Analysis5)数据作对比,在不同风速、水深的海域都获得了基本一致的结果,超过半数区域的海浪周期误差在5%以内,初步证明了星载光子计数激光雷达观测成果用于海浪要素计算的可行性。

     

    Abstract: The Advanced Topographic Laser Altimeter System (ATLAS) carried on the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) launched by NASA in 2018 is the only spaceborne photon counting lidar in the world so far. It has a high orbital spatial sampling rate, which makes it possible to detect ocean wave elements by remote sensing. The premise of photon counting lidar for wave detection is to accurately extract signal photons from the sea surface and determine the instantaneous sea surface profile. Up to now, there are few reports on the detection of sea surface morphology and wave elements by spaceborne photon counting lidar, and there is also a lack of extraction methods specifically for sea surface signal photons. Based on the distribution characteristics of signal photons on the sea surface, a new signal extraction algorithm is proposed in this paper. First, the rough denoising of sea surface echo photons is completed through histogram statistics and adaptive threshold selection; Then, based on the lidar spot size and sea surface fluctuation characteristics, an appropriate search neighborhood is selected to calculate the density of signal points and noise points. The point density difference is used to classify signal photons and noise photons; Finally, the backscattered noise photons with higher density are further removed by Gaussian function fitting, and the signal photons reflected from the sea surface are obtained. Using the above algorithm, the sea surface signal photons and instantaneous sea surface profiles of seven different sea state regions in the Pacific Ocean are extracted, and the peak wavelength and peak wave period of local ocean waves are further calculated. Comparing the calculation results with the global atmospheric reanalysis ECMWF Re-Analysis5 (ERA5) data of the European Centre for Medium-Range Weather Forecasts (ECMWF) during the same period, the results are basically consistent in sea areas with different wind speeds and water depths. As a result, the relative error of the wave period in more than half of the areas is within 5%, which preliminarily proves the feasibility of calculating ocean wave elements with the spaceborne photon counting lidar observation results.

     

/

返回文章
返回