Integrating machine learning and physical models for reconstructing sound speed profiles in mesoscale eddies

Hongchen LI; Ming LI; Penghao WANG; Kefeng MAO; Yuhang ZHU; Yuhang LIU

doi:10.1007/s00343-025-5263-1

Your Location：

Home >

Browse articles >

Integrating machine learning and physical models for reconstructing sound speed profiles in mesoscale eddies

Physics | Updated：2026-08-27

- Integrating machine learning and physical models for reconstructing sound speed profiles in mesoscale eddies
- Journal of Oceanology and Limnology Vol. 44, Issue 4, Pages: 1323-1341(2026)
- Affiliations：
  
  1.College of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, China
  2.College of Frontier Interdisciplinary Sciences, National University of Defense Technology, Nanjing 211101, China
  3.College of Marine Science and Technology, Zhejiang Ocean University, Zhoushan 316022, China
  4.College of Marine Science and Ecological Environment, Shanghai Ocean University, Shanghai 201306, China
  5.College of Marine Technology and Surveying, Jiangsu Ocean University, Lianyungang 222000, China
- Author bio：
  
  mingli152@163.com
- Funds：
- DOI：10.1007/s00343-025-5263-1
  CLC：
- Received：20 August 2025，
  
  Accepted：11 October 2025，
  
  Online First：26 May 2026，
  
  Published：01 July 2026
- Accepted：
Scan QR Code
LI Hongchen,LI Ming,WANG Penghao,et al.Integrating machine learning and physical models for reconstructing sound speed profiles in mesoscale eddies[J].Journal of Oceanology and Limnology,2026,44(04):1323-1341.
DOI：

LI Hongchen,LI Ming,WANG Penghao,et al.Integrating machine learning and physical models for reconstructing sound speed profiles in mesoscale eddies[J].Journal of Oceanology and Limnology,2026,44(04):1323-1341. DOI： 10.1007/s00343-025-5263-1.

摘要

Abstract

Mesoscale eddies significantly affect sound propagation

yet their complex internal sound speed profiles suffer from substantial reconstruction errors

and only few studies specifically addressed this reconstruction. To bridge this gap

we used multi-source satellite data and Argo profiles to identify eddies and build a temperature-salinity-sound speed dataset. Sea surface parameters (temperature

height anomalies

salinity

density) and Argo density serve as inputs for a random forest (RF) algorithm to learn the surface-to-underwater sound speed mapping. Concurrently

a unified eddy dynamic model reconstructs the internal density field. By combining these environmental parameters with the reconstructed density generates eddy sound speed profiles

an integrated PIRF-DEN model that merges machine learning and physical modeling was established. Evaluations demonstrate the superiority of the model. By incorporating density input

the reconstruction accuracy was significantly improved

the mean absolute error (MAE) and root mean square error (RMSE) were reduced to 0.83 and 1.39 m/s

respectively

which is 87.3% and 83.7% less than that of the sEOFr method. Integration of the eddy model effectively characterized the vertical density structure

whose constraint lowered the overfitting risk and enhanced the accuracy and stability over sEOFr

sEOF-RF

and RF models. Propagation loss calculations using the reconstructed sound speed showed high correlation (coefficient: 0.77) with measured data

further confirming its reliability.

关键词

Keywords

references

Alexander P , Duncan A , Bose N et al . 2016 . Modelling acoustic propagation beneath Antarctic sea ice using measured environmental parameters . Deep Sea Research Part II: Topical Studies in Oceanography , 131 : 84 - 95 , https://doi.org/10.1016/j.dsr2.2016.04.026 https://doi.org/10.1016/j.dsr2.2016.04.026 .

Bao S L . 2021 . Quality Control of Satellite Salinity Products and the Reconstruction of 3D Temperature and Salinity Fields in Tropical India Ocean Based on Multi-Source Information . National University of Defense Technology, Changsha, China , https://doi.org/10.27052/d.cnki.gzjgu.2021.000023. https://doi.org/10.27052/d.cnki.gzjgu.2021.000023. (in Chinese with English abstract)

Chen C , Ma Y L , Liu Y . 2018a . Reconstructing sound speed profiles worldwide with sea surface data . Applied Ocean Research , 77 : 26 - 33 , https://doi.org/10.1016/j.apor.2018.05.002 https://doi.org/10.1016/j.apor.2018.05.002 .

Chen C , Yang K D , Duan R et al . 2017 . Acoustic propagation analysis with a sound speed feature model in the front area of Kuroshio Extension . Applied Ocean Research , 68 : 1 - 10 , https://doi.org/10.1016/j.apor.2017.08.001 https://doi.org/10.1016/j.apor.2017.08.001 .

Chen C T , Millero F J . 1977 . Speed of sound in seawater at high pressures . The Journal of the Acoustical Society of America , 62 ( 5 ): 1129 - 1135 , https://doi.org/10.1121/1.381646 https://doi.org/10.1121/1.381646 .

Chen W , Zhang Y C , Liu Y Y et al . 2022a . Observation of a mesoscale warm eddy impacts acoustic propagation in the slope of the South China Sea. Frontiers in Marine Science , 9 : 1086799 , https://doi.org/10.3389/fmars.2022.1086799 https://doi.org/10.3389/fmars.2022.1086799 .

Chen W , Zhang Y C , Liu Y Y et al . 2022b . Parametric model for eddies-induced sound speed anomaly in five active mesoscale eddy regions . Journal of Geophysical Research: Oceans , 127 ( 8 ): e2022 JC 018408 , https://doi.org/10.1029/2022JC018408 https://doi.org/10.1029/2022JC018408 .

Chen X , Hu D , Mao K F et al . 2018b . Detailed investigation of the three-dimensional structure of a mesoscale cold eddy in the Kuroshio extension region . Journal of Operational Oceanography , 11 ( 2 ): 87 - 99 , https://doi.org/10.1080/1755876X.2018.1505069 https://doi.org/10.1080/1755876X.2018.1505069 .

Feng X , Tian T , Zhou M Z et al . 2024 . Sound speed inversion based on multi-source ocean remote sensing observations and machine learning . Remote Sensing , 16 ( 5 ): 814 , https://doi.org/10.3390/rs16050814 https://doi.org/10.3390/rs16050814 .

Fox D N , Teague W J , Barron C N et al . 2002 . The modular ocean data assimilation system (MODAS) . Journal of Atmospheric and Oceanic Technology , 19 ( 2 ): 240 - 252 , https://doi.org/10.1175/1520-0426(2002)019<0240:TMODAS>2.0.CO;2. https://doi.org/10.1175/1520-0426(2002)019<0240:TMODAS>2.0.CO;2.

Lapeyre G , Klein P . 2006 . Dynamics of the upper oceanic layers in terms of surface quasigeostrophy theory . Journal of Physical Oceanography , 36 ( 2 ): 165 - 176 , https://doi.org/10.1175/JPO2840.1 https://doi.org/10.1175/JPO2840.1 .

LeBlanc L R , Middleton F H . 1980 . An underwater acoustic sound velocity data model . The Journal of the Acoustical Society of America , 67 ( 6 ): 2055 - 2062 , https://doi.org/10.1121/1.384448 https://doi.org/10.1121/1.384448 .

Le Vu B , Stegner A , Arsouze T . 2018 . Angular momentum eddy detection and tracking algorithm (AMEDA) and its application to coastal eddy formation . Journal of Atmospheric and Oceanic Technology , 35 ( 4 ): 739 - 762 , https://doi.org/10.1175/JTECH-D-17-0010.1 https://doi.org/10.1175/JTECH-D-17-0010.1 .

Li H C , Li M , Chen X et al . 2025 . Fast reconstruction of sound speed profile based on machine learning and empirical orthogonal function method . Oceanologia et Limnologia Sinica , 56 ( 1 ): 101 - 111 , https://doi.org/10.11693/hyhz20240400096. https://doi.org/10.11693/hyhz20240400096. (in Chinese with English abstract)

Li H P , QU K , Zhou J B . 2021 . Reconstructing sound speed profile from remote sensing data: nonlinear inversion based on self-organizing map . IEEE Access , 9 : 109754 - 109762 , https://doi.org/10.1109/access.2021.3102608 https://doi.org/10.1109/access.2021.3102608 .

Li M , Liu Y H , Sun Y Y et al . 2024 . Quantitative analysis and prediction of the sound field convergence zone in mesoscale eddy environment based on data mining methods . Acta Oceanologica Sinica , 43 ( 5 ): 110 - 120 , https://doi.org/10.1007/s13131-024-2328-5 https://doi.org/10.1007/s13131-024-2328-5 .

Li Q Q , Li H L , Cao S L et al . 2022 . Inversion of the full-depth sound speed profile based on remote sensing data and surface sound speed . Haiyang Xuebao , 44 ( 12 ): 84 - 94 , https://doi.org/10.12284/hyxb2022149. https://doi.org/10.12284/hyxb2022149. (in Chinese with English abstract)

Lin Y T , Lynch J . 2017 . Three-dimensional sound propagation and scattering in an ocean with surface and internal waves over range-dependent seafloor . The Journal of the Acoustical Society of America , 141 ( S5 ): 3753 , https://doi.org/10.1121/1.4988280 https://doi.org/10.1121/1.4988280 .

Liu C . 2020 . Sound Speed Profile Inversion Based on Empirical Orthogonal Function Method and its Influence on Acoustic Propagation Characteristics . National University of Defense Technology, Changsha, China , https://doi.org/10.27052/d.cnki.gzjgu.2020.000842. https://doi.org/10.27052/d.cnki.gzjgu.2020.000842. (in Chinese with English abstract)

Liu L , Peng S Q , Huang R X . 2017 . Reconstruction of ocean's interior from observed sea surface information . Journal of Geophysical Research: Oceans , 122 ( 2 ): 1042 - 1056 , https://doi.org/10.1002/2016JC011927 https://doi.org/10.1002/2016JC011927 .

Liu Y F , Wang Z J , Zhao S . 2020 . Layered-EOFs based adaptive reconstruction of sound velocity profile in multi-beam sounding . Technical Acoustics , 39 ( 3 ): 372 - 378 , https://doi.org/10.16300/j.cnki.1000-3630.2020.03.020. https://doi.org/10.16300/j.cnki.1000-3630.2020.03.020. (in Chinese with English abstract)

Liu Y Y , Chen W , Chen Y et al . 2024a . Research progress on methods for constructing underwater sound speed profile . Information Countermeasure Technology , 3 ( 5 ): 1 - 18 , https://doi.org/10.12399/j.issn.2097-163x.2024.05.001. https://doi.org/10.12399/j.issn.2097-163x.2024.05.001. (in Chinese with English abstract)

Liu Y Y , Chen Y , Chen W et al . 2024b . Inversion of sound speed profile in the Luzon strait by combining single empirical orthogonal function and generalized regression neural network. IEEE Geoscience and Remote Sensing Letters , 21 : 1502405 , https://doi.org/10.1109/LGRS.2024.3379200 https://doi.org/10.1109/LGRS.2024.3379200 .

Liu Y Y , Chen Y , Meng Z et al . 2023 . Performance of single empirical orthogonal function regression method in global sound speed profile inversion and sound field prediction. Applied Ocean Research , 136 : 103598 , https://doi.org/10.1016/j.apor.2023.103598 https://doi.org/10.1016/j.apor.2023.103598 .

North G R , Bell T L , Cahalan R F et al . 1982 . Sampling errors in the estimation of empirical orthogonal functions . Monthly Weather Review , 110 ( 7 ): 699 - 706 . https://doi.org/10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2. https://doi.org/10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2.

Ou Z Y , Qu K , Wang Y F et al . 2022 . Estimating sound speed profile by combining satellite data with in situ sea surface observations . Electronics , 11 ( 20 ): 3271 , https://doi.org/10.3390/electronics11203271 https://doi.org/10.3390/electronics11203271 .

Park J C , Kennedy R M . 1996 . Remote sensing of ocean sound speed profiles by a perceptron neural network . IEEE Journal of Oceanic Engineering , 21 ( 2 ): 216 - 224 , https://doi.org/10.1109/48.486796 https://doi.org/10.1109/48.486796 .

Scott R B , Wang F M . 2005 . Direct evidence of an oceanic inverse kinetic energy cascade from satellite altimetry . Journal of Physical Oceanography , 35 ( 9 ): 1650 - 1666 , https://doi.org/10.1175/JPO2771.1 https://doi.org/10.1175/JPO2771.1 .

Wang J B , Flierl G R , LaCasce J H et al . 2013 . Reconstructing the ocean's interior from surface data . Journal of Physical Oceanography , 43 ( 8 ): 1611 - 1626 , https://doi.org/10.1175/JPO-D-12-0204.1 https://doi.org/10.1175/JPO-D-12-0204.1 .

Wang P H , Mao K F , Chen X et al . 2022 . The three-dimensional structure of the mesoscale eddy in the Kuroshio extension region obtained from three datasets . Journal of Marine Science and Engineering , 10 ( 11 ): 1754 , https://doi.org/10.3390/jmse10111754 https://doi.org/10.3390/jmse10111754 .

Wang P H , Mao K F , Chen X et al . 2025 . Three-dimensional structure reconstruction of ocean mesoscale eddies based on physical process modeling and data-driven machine learning. Ocean Modelling , 196 : 102558 , https://doi.org/10.1016/j.ocemod.2025.102558 https://doi.org/10.1016/j.ocemod.2025.102558 .

White W B , McCreary J P . 1976 . On the formation of the Kuroshio meander and its relationship to the large-scale ocean circulation . Deep Sea Research and Oceanographic Abstracts , 23 ( 1 ): 33 - 47 , https://doi.org/10.1016/0011-7471(76)90806-8 https://doi.org/10.1016/0011-7471(76)90806-8 .

Wong G S K , Zhu S M . 1995 . Speed of sound in seawater as a function of salinity, temperature, and pressure . The Journal of the Acoustical Society of America , 97 ( 3 ): 1732 - 1736 , https://doi.org/10.1121/1.413048 https://doi.org/10.1121/1.413048 .

Wu S L , Li Z L , Qin J X et al . 2022 . The effects of sound speed profile to the convergence zone in deep water . Journal of Marine Science and Engineering , 10 ( 3 ): 424 , https://doi.org/10.3390/jmse10030424 https://doi.org/10.3390/jmse10030424 .

Yan H Q . 2021 . Fast Dynamical-Statistical Reconstruction of Three-Dimensional Ocean Data: a Research on the Framework and Key Techniques . National University of Defense Technology, Changsha, China , https://doi.org/10.27052/d.cnki.gzjgu.2021.000441. https://doi.org/10.27052/d.cnki.gzjgu.2021.000441. (in Chinese with English abstract)

Yasuda I , Okuda K , Hirai M . 1992 . Evolution of a Kuroshio warm-core ring—variability of the hydrographic structure . Deep Sea Research Part A. Oceanographic Research Papers , 39 ( S1 ): S131 - S161 , https://doi.org/10.1016/S0198-0149(11)80009-9 https://doi.org/10.1016/S0198-0149(11)80009-9 .

Zhang Z G . 2014 . Mesoscale Eddy . Ocean University of China, China .

Zhao Y , Xu P , Li G M et al . 2024 . Reconstructing the sound speed profile of South China Sea using remote sensing data and long short-term memory neural networks. Frontiers in Marine Science , 11 : 1375766 , https://doi.org/10.3389/fmars.2024.1375766 https://doi.org/10.3389/fmars.2024.1375766 .

Zhu F Q , Zhang H G , Qu K . 2021 . Influence of mesoscale warm eddies on sound propagation in the northeastern South China Sea . Journal of Harbin Engineering University , 42 ( 10 ): 1496 - 1502 , https://doi.org/10.11990/jheu.202007077. https://doi.org/10.11990/jheu.202007077. (in Chinese with English abstract)

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

High-precision classification of benthic habitat sediments in shallow waters of islands by multi-source data

Enhancing reproductive efficiency of mud crab Scylla paramamosain in RAS: effects of temperature and eyestalk ablation on ovarian development, lipid metabolism, and spawning rate

Fungal diversity associated with Ulva pertusa and its response to cAMP and light

Mechanism of palmelloid formation induced by polystyrene nanoplastics in Chlamydomonas reinhardtii: a multi-omics analysis

Cloning, tissue expression, and functional analysis of PfTDRD5, PfTDRD7-like, and PfTDRD9-like genes in pearl oyster Pinctada fucata

Related Author

Qiuhua TANG

Ningning LI

Yujie ZHANG

Zhipeng DONG

Yongling ZHENG

Jingjing BAO

Jingyu ZHANG

Ningning LI

Related Institution

First Institute of Oceanography, Ministry of Natural Resources

State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing

College of Mapping and Spatial Information, Shandong University of Science and Technology

Third Institute of Oceanography, Ministry of Natural Resources

Address：Editorial Office of JOL, 88 Haijun Rd. Qingdao, 266000, China
Tel：+86-532-82898754 Email：jol@qdio.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰