测绘学报 ›› 2024, Vol. 53 ›› Issue (8): 1517-1530.doi: 10.11947/j.AGCS.2024.20230558

• 大地测量与导航 • 上一篇    下一篇

移去-恢复法在基于重力异常的海底地形解析迭代算法中的应用

安邦1,2(), 于瑶瑶1,3(), 徐焕1,2, 于锦海1,2, 田彧玮1,2   

  1. 1.中国科学院大学地球与行星科学学院,北京 100049
    2.中国科学院计算地球动力学重点实验室,北京 100049
    3.北京市遥感信息研究所,北京 100124
  • 收稿日期:2023-12-01 发布日期:2024-09-25
  • 通讯作者: 于瑶瑶 E-mail:anbang19@mails.ucas.ac.cn;anbang19@mails.ucas.ac.cn;yu.yaoyao9@outlook.com
  • 作者简介:安邦(1998—),男,博士生,研究方向为重力场计算与应用。E-mail:anbang19@mails.ucas.ac.cn
  • 基金资助:
    国家自然科学基金(42274010)

Application of remove-restore method in iterative algorithm for submarine terrain analysis based on gravity anomaly data

Bang AN1,2(), Yaoyao YU1,3(), Huan XU1,2, Jinhai YU1,2, Yuwei TIAN1,2   

  1. 1.College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
    2.Key Laboratory of Computational Geodynamics, Chinese Academy of Sciences, Beijing 100049, China
    3.Capital Institute of Geographic Information, Beijing 100124, China
  • Received:2023-12-01 Published:2024-09-25
  • Contact: Yaoyao YU E-mail:anbang19@mails.ucas.ac.cn;anbang19@mails.ucas.ac.cn;yu.yaoyao9@outlook.com
  • About author:AN Bang (1998—), male, PhD candidate, majors in gravity field calculation and application. E-mail: anbang19@mails.ucas.ac.cn
  • Supported by:
    The National Natural Science Foundation of China(42274010)

摘要:

高精度的海底地形图对于地球科学研究有着重要意义,由于直接的海深测量难以进行高密度全球覆盖,因此利用重力数据预测海底地形成为非常有效的方法。当前重力预测海底地形的方法有频率导纳法、重力地质法、解析迭代法等,但都存在一定的问题。本文针对解析迭代算法中难以处理的远区地形和深部密度异常的低频影响,应用移去-恢复方法对这些低频影响进行剔除。结合现有海深模型,通过仿真试验,讨论了分离低频影响的方法与相应的精度,结论如下:对于在移去-恢复方法中使用的初始海深模型,在海面上剔除掉重力异常的低频影响后,反演解算得到的海深结果均较初始海深模型的精度有所提高,验证了移去-恢复方法在利用重力异常数据反演海底地形的有效性。本文选取了中国南海12.87°N—13.87°N,113.65°E—114.65°E区域,利用移去-恢复方法解算了其下方的海底地形,将解算结果与NGDC的实测海深数据进行比较,预测结果的均方根误差为90.3 m,相对误差可达2.11%。这为利用高精度重力数据,通过解析迭代算法改善船测稀疏区域的海深数据提供了可能。

关键词: 海底地形, 重力异常, 解析迭代算法, 移去-恢复方法

Abstract:

High-precision seafloor topography maps are of great significance in earth science research. Due to the difficulties of measuring high-resolution global coverage of bathymetry directly, a large-scale prediction of seafloor topography relies on gravity data mostly. So, various methods are employed for predicting seafloor topography from gravity data, such as the frequency-domain polarization method, geological-gravity method and iterative analytical method. However, there are some drawbacks in these methods. In this paper, remove-restore method is applied to deal with the issues related to the far-field influences in the iterative analytical method. By combining existing bathymetric models, the simulated experiments are designed to discuss separation methods for the far-field influences and their accuracies are also assessed in inverting seafloor topography. Ultimately, the effectiveness of the remove-restore method is validated in improving seafloor topography by applying to the region of 13.85°N—14.85°N latitude and 117.25°E—118.25°E longitude in the South China Sea. Compared with actual bathymetric data from NGDC, it is concluded that the root mean square error is 90.3 meters and relative error is up to 2.11%. This provides a potential means for improving sparse bathymetric data in ship-measuring areas with high-precision gravity data and the analytical iterative algorithm.

Key words: seafloor topography, gravity anomaly, analytical iterative algorithm, remove-restore method

中图分类号: