Recovering Gravity from Satellite Altimetry Data Using Deep Learning Network

doi:10.1109/TGRS.2023.3280261

IOCAS-IR

	Recovering Gravity from Satellite Altimetry Data Using Deep Learning Network
	Zhu, Chengcheng 1; Yang, Lei 2,3,4; Bian, Hongwei 5; Li, Houpu 5; Guo, Jinyun 6; Liu, Na 7; Lin, Lina 7
	2023
发表期刊	IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
ISSN	0196-2892
卷号	61 页码:11
通讯作者	Yang, Lei(leiyang@fio.org.cn)
摘要	The satellite altimetry missions could measure high-accuracy sea surface heights (SSHs) that can be used to recover the marine gravity field. Traditional methods for estimating the marine gravity field from SSHs all rely on approximate physical correlations between SSHs and gravity, which may neglect nature's complex nonlinearity. This work presents a new deep network-based method to recover the gravity anomaly. This new method uses a multichannel convolutional neural network (MCCNN) architecture to capture the nonlinear features between ship-borne gravity and a group of input parameters including deflections of the vertical (DOVs), submarine topography, and the geo-locations. To validate the gravity, ship-borne gravity anomalies on the two independent cruises were not used in the deep learning process. For comparison, we also estimated the gravity using the traditional inverse Vening Meinesz (IVM) method. Our results indicate that the MCCNN method can derive high-quality marine gravity anomalies. The assessments using 1-mGal-accuracy ship-borne gravity anomalies show that the average accuracy for gravity from the MCCNN method is higher than 3 mGal and this method achieves 0.05-0.50 mGal improvement over benchmark methods IVM. Assessed by marine gravity anomaly models with the accuracy of 1-2 mGal, the MCCNN method has been shown to improve the accuracy of gravity by at least 4%. Comparisons with the IVM results show that improvements in the MCCNN method were mainly in wavelengths between 8 and 100 km due to the use of bathymetry. The results show that our deep learning method maintains good performance and is promising for gravity recovery.
关键词	Gravity Deep learning Sea measurements Satellites Underwater vehicles Training Data models gravity anomaly multichannel convolutional neural network (MCCNN) satellite altimetry Index Terms submarine topography
DOI	10.1109/TGRS.2023.3280261
收录类别	SCI
语种	英语
资助项目	National Science Foundation for Outstanding Young Scholars[42122025] ; National Natural Science Foundation of China[41876222] ; Shandong Provincial Natural Science Foundation[ZR2022QD025]
WOS研究方向	Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology
WOS类目	Geochemistry & Geophysics ; Engineering, Electrical & Electronic ; Remote Sensing ; Imaging Science & Photographic Technology
WOS记录号	WOS:001012873600008
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS关键词	TOPEX/POSEIDON ALTIMETRY ; GEOSAT ; SEASAT ; ERS-1 ; MODEL
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.qdio.ac.cn/handle/337002/182362
专题	中国科学院海洋研究所
通讯作者	Yang, Lei
作者单位	1.Shandong Jianzhu Univ, Sch Surveying & Geoinformat, Jinan 250101, Peoples R China 2.Minist Nat Resources, Inst Oceanog 1, Qingdao 266061, Peoples R China 3.Chinese Acad Sci, Inst Oceanol, Qingdao 266071, Peoples R China 4.Univ Chinese Acad Sci, Beijing 100864, Peoples R China 5.Naval Univ Engn, Coll Elect Engn, Wuhan 430033, Peoples R China 6.Shandong Univ Sci & Technol, Coll Geodesy & Geomat, Qingdao 266590, Peoples R China 7.Minist Nat Resources, Inst Oceanog 1, Qingdao 266061, Peoples R China
通讯作者单位	中国科学院海洋研究所
推荐引用方式 GB/T 7714	Zhu, Chengcheng,Yang, Lei,Bian, Hongwei,et al. Recovering Gravity from Satellite Altimetry Data Using Deep Learning Network[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2023,61:11.
APA	Zhu, Chengcheng.,Yang, Lei.,Bian, Hongwei.,Li, Houpu.,Guo, Jinyun.,...&Lin, Lina.(2023).Recovering Gravity from Satellite Altimetry Data Using Deep Learning Network.IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,61,11.
MLA	Zhu, Chengcheng,et al."Recovering Gravity from Satellite Altimetry Data Using Deep Learning Network".IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 61(2023):11.