ISSN 2096-5508

e-ISSN 2523-3521

CN 37-1518/P

Impact factor：1.6

Submit and Review

 

JOL at Springer

 

 

 

 

  • Current Issue
  • Volumes and Issues
  • Online First

  • Physics

    Abstract：Physics-informed neural networks (PINNs), as a novel artificial intelligence method for solving partial differential equations, are applicable to solve both forward and inverse problems. This study evaluates the performance of PINNs in solving the temperature diffusion equation of the seawater across six scenarios, including forward and inverse problems under three different boundary conditions. Results demonstrate that PINNs achieved consistently higher accuracy with the Dirichlet and Neumann boundary conditions compared to the Robin boundary condition for both forward and inverse problems. Inaccurate weighting of terms in the loss function can reduce model accuracy. Additionally, the sensitivity of model performance to the positioning of sampling points varied between different boundary conditions. In particular, the model under the Dirichlet boundary condition exhibited superior robustness to variations in point positions during the solutions of inverse problems. In contrast, for the Neumann and Robin boundary conditions, accuracy declines when points were sampled from identical positions or at the same time. Subsequently, the Argo observations were used to reconstruct the vertical diffusion of seawater temperature in the north-central Pacific for the applicability of PINNs in the real ocean. The PINNs successfully captured the vertical diffusion characteristics of seawater temperature, reflected the seasonal changes of vertical temperature under different topographic conditions, and revealed the influence of topography on the temperature diffusion coefficient. The PINNs were proved effective in solving the temperature diffusion equation of seawater with limited data, providing a promising technique for simulating or predicting ocean phenomena using sparse observations.  

    Lei HAN, Changming DONG, Yuli LIU, Huarong XIE, Hongchun ZHANG, Weijun ZHU

    |

  • Physics

    Abstract：The prediction of sea surface partial pressure of carbon dioxide (pCO2) in the South China Sea is crucial for understanding the region’s contribution to the global carbon budget and its interactions with climate change. We applied the Spatiotemporal Convolutional Long Short-Term Memory (ST-ConvLSTM) model, integrating key environmental factors including sea surface temperature (SST), sea surface salinity (SSS), and chlorophyll a (Chl a), to predict and analyze sea surface pCO2 in the South China Sea. The model demonstrated high accuracy in short-term predictions (1 month), with a mean absolute error (MAE) of 0.394, a root mean square error (RMSE) of 0.659, and a coefficient of determination (R2) of 0.998. For long-term predictions (12 months), the model maintained its predictive capability, with an MAE of 0.667, RMSE of 1.255, and R2 of 0.994. Feature importance analysis revealed that sea surface pCO2 and SST were the main drivers of the model’s predictions, whereas Chl a and SSS had relatively minor impacts. The model’s generalization ability was further validated in the northwest Pacific Ocean and tropical Pacific Ocean, where it successfully captured the spatiotemporal variation in pCO2 with small prediction errors. The ST-ConvLSTM model provides an efficient and accurate tool for forecasting and analyzing sea surface pCO2 in the South China Sea, offering new insights into global carbon cycling and climate change. This study demonstrates the potential of deep learning in marine science and provides a significant technical support for global changes and marine ecosystem research.  

    Shuang LI, Yu GAO, Jiannan GAO, Yaqi ZHAO, Peng HAO, Jinbao SONG, Chengcheng YU

    |

  • Physics

    Abstract：Bathymetric measurement of shallow water is of fundamental importance to coastal environment research and resource management. However, there are still great challenges in estimating water depth using satellite observations in turbid coastal waters. In this paper, we developed a physics-enhanced deep neural network to estimate bathymetry of highly turbid waters of the Changjiang (Yangtze) River estuary from dual-polarized synthetic aperture radar (SAR) images. Sentinel-1A/B SAR images with a spatial resolution of 20 m×22 m were collected and matched with water depth data from nautical charts during 2017–2023. For the input parameters of the model, in addition to the normalized radar backscatter cross section (NRCS) at single polarization and incidence angle, the impacts of both polarimetric characteristics and physical environmental factors on model performance were discussed in detail. Results of feature importance analysis and sensitivity experiments indicate that the polarization ratio and NRCS after removing the influence of background sea surface wind field make significant contributions to the bathymetry retrieval model. The root mean square error (RMSE) of SAR derived water depth decreases from 1.44 to 0.78 m within 0–30-m depth, and the mean relative error (MRE) is reduced from 15.6% to 8.6%. Compared with other machine learning models such as ResNet, XGBoost, and Random Forest, the MRE is reduced by 3.9%, 5.7%, and 7.4%, respectively. The spatial distribution of SAR derived water depth also exhibits a high degree of consistency with observations, demonstrating the great potential of the model in estimating the depth of turbid shallow waters.  

    Tian MA, Qing XU, Xiaobin YIN, Yan LI, Letian LÜ, Kaiguo FAN

    |

  • Physics

    Abstract：The study of mesoscale eddies is generally categorized in Eulerian or Lagrangian frameworks. We employed the eddy identification techniques in both frameworks in the South China Sea (SCS), examining the differential characteristics of mesoscale eddies ascertained through each approach, and attempting to identify factors influencing eddy lifetime. The findings suggest that eddies identified via the sea surface height (SSH) method in the Eulerian framework typically have larger spatial extents compared to those identified using the Lagrangian Average Vorticity Deviation (LAVD) method. The latter is characterized by a greater number of vortices with smaller average values of characteristic parameters. SSH eddies exhibited more remarked seasonal variations than LAVD vortices, and the seasonal variations of their respective cyclonic and anticyclonic eddies showed opposite trends. Analysis in both frameworks indicates that eddy lifetime is positively correlated with various eddy characteristic parameters, including radius, vorticity, kinetic energy, amplitude, EKE/MKE (ratio of boundary to spatial mean kinetic energy), and U/c (max rotation speed to mean propagation speed ratio). A subsequent comparison between SSH eddies with LAVD cores (SSH eddy with LAVD vortex inside) and those without reveals a greater likelihood of extended lifetime in the former. Compared to the characteristic parameters of eddies, the presence of LAVD cores emerges as a critical factor in determining the lifetime of SSH eddies.  

    Zekai CHEN, Yifan LIU, Qiong XIA

    |
Loading...
More >
  • Cover Articles
  • Special Issues
  • Virtual Issues
  • Meetings
  • Videos

Citation Ranking

>

Browse Ranking

>

Download ranking

>
0