Sea surface temperature (SST) is an important physical factor in the interaction between the ocean and the atmosphere. Accurate monitoring and prediction of the temporal and spatial distribution of SST are of great significance in dealing with climate change, disaster prevention, disaster reduction, and marine ecological protection. This study establishes a prediction model of sea surface temperature for the next five days in the East China Sea using long-term and short-term memory neural networks (LSTM). It investigates the influence of different parameters on prediction accuracy. The sensitivity experiment results show that, based on the same training data, the length of the input data of the LSTM model can improve the model’s prediction performance to a certain extent. However, no obvious positive correlation is observed between the increase in the input data length and the improvement of the model’s prediction accuracy. On the contrary, the LSTM model’s performance decreases with the prediction length increase. Furthermore, the single-point prediction results of the LSTM model for the estuary of the Yangtze River, Kuroshio, and the Pacific Ocean are accurate. In particular, the prediction results of the point in the Pacific Ocean are the most accurate at the selected four points, with an RMSE of 0.0698 °C and an $R^2$ of 99.95%. At the same time, the model in the Pacific region is migrated to the East China Sea. The model was found to have good mobility and can well represent the long-term and seasonal trends of SST in the East China Sea.

Deep learning methods have excellent prospects for application in wave forecasting research. This study employed the convolutional LSTM (ConvLSTM) algorithm to predict the South China Sea (SCS) significant wave height (SWH). Three prediction models were established to investigate the influences of setting different parameters and using multiple training data on the forecasting effects. Compared with the SWH data from the China–France Ocean Satellite (CFOSAT), the SWH of WAVEWATCH III (WWIII) from the pacific islands ocean observing system are accurate enough to be used as training data for the ConvLSTM-based SWH prediction model. Model A was preliminarily established by only using the SWH from WWIII as the training data, and 20 sensitivity experiments were carried out to investigate the influences of different parameter settings on the forecasting effect of Model A. The experimental results showed that Model A has the best forecasting effect when using three years of training data and three hourly input data. With the same parameter settings as the best prediction performance Model A, Model B and C were also established by using more different training data. Model B used the wind shear velocity and SWH as training and input data. When making a 24-h SWH forecast, compared with Model A, the root mean square error (RMSE) of Model B is decreased by 17.6%, the correlation coefficient (CC) is increased by 2.90%, and the mean absolute percentage error (MAPE) is reduced by 12.2%. Model C used the SWH, wind shear velocity, wind and wave direction as training and input data. When making a 24-h SWH forecast, compared with Model A, the RMSE of Model C decreased by 19.0%, the CC increased by 2.65%, and the MAPE decreased by 14.8%. As the performance of the ConvLSTM-based prediction model mainly rely on the SWH training data. All the ConvLSTM-based prediction models show a greater RMSE in the nearshore area than that in the deep area of SCS and also show a greater RMSE during the period of typhoon transit than that without typhoon. Considering the wind shear velocity, wind, and wave direction also used as training data will improve the performance of SWH prediction.

The present work proposes a prediction model of significant wave height (SWH) and average wave period (APD) based on variational mode decomposition (VMD), temporal convolutional networks (TCNs), and long short-term memory (LSTM) networks. The wave sequence features were obtained using VMD technology based on the wave data from the National Data Buoy Center. Then the SWH and APD prediction models were established using TCNs, LSTM, and Bayesian hyperparameter optimization. The VMD–TCN–LSTM model was compared with the VMD–LSTM (without TCN cells) and LSTM (without VMD and TCN cells) models. The VMD–TCN–LSTM model has significant superiority and shows robustness and generality in different buoy prediction experiments. In the 3 h wave forecasts, VMD primarily improved the model performance, while the TCN had less of an influence. In the 12, 24, and 48 h wave forecasts, both VMD and TCNs improved the model performance. The contribution of the TCN to the improvement of the prediction result determination coefficient gradually increased as the forecasting length increased. In the 48 h SWH forecasts, the VMD and TCN improved the determination coefficient by 132.5 % and 36.8 %, respectively. In the 48 h APD forecasts, the VMD and TCN improved the determination coefficient by 119.7 % and 40.9 %, respectively.

Storm surge forecasting presents a significant challenge for coastal resilience, particularly in typhoon-prone regions such as southeastern China, where compound flooding events lead to substantial socioeconomic losses. Although artificial intelligence (AI) models have shown strong potential in storm surge prediction, their inherent “black-box” nature limits both their interpretability and operational trust. In this study, we integrate a Vision Transformer (ViT) model with an explainable AI (XAI) method—specifically, Shapley value analysis (SHAP)—to develop an interpretable, high-performance storm surge forecasting framework. The baseline ViT model demonstrates excellent predictive skill, achieving spatiotemporal correlation coefficients exceeding 0.90 over a 12 h lead time. However, it exhibits systematic underestimations in topographically complex regions, such as semi-enclosed bays (e.g., up to 0.06 m). SHAP analysis reveals that the model primarily relies on the autocorrelation of historical surge levels rather than external wind forcing—contrary to the conventional physical understanding of storm surge dynamics. Guided by these insights, we introduce the surge time difference (ΔZ/Δt) as an explicit input feature to enhance the model’s physical representation. This modification yields substantial improvements: during the critical first hour of forecasting—a key window for disaster mitigation—the RMSE is reduced from 0.01 m to 0.005 m, while the correlation coefficient increases from 0.92 to 0.98. This study bridges the gap between data-driven forecasting and physical interpretability, offering a transparent and trustworthy framework for next-generation intelligent storm surge prediction.

The present work proposes a prediction model of significant wave height (SWH) and average wave period (APD) based on variational mode decomposition (VMD), temporal convolutional networks (TCN), and long short-term memory (LSTM) networks. The wave sequence features were obtained using VMD technology based on the wave data from the National Data Buoy Center. Then the SWH and APD prediction models were established using TCN, LSTM, and Bayesian hyperparameter optimization. The VMD-TCN-LSTM model was compared with the VMD-LSTM (without TCN cells) and LSTM (without VMD and TCN cells) models. The VMD-TCN-LSTM model has significant superiority and shows robustness and generality in different buoy prediction experiments. In the 3-hour wave forecasts, VMD primarily improved the model performance, while the TCN had less influence. In the 12-, 24-, and 48-hour wave forecasts, both VMD and TCN improved the model performance. The contribution of the TCN to the improvement of the prediction result determination coefficient gradually increased as the forecasting length increased. In the 48-hour SWH forecasts, the VMD and TCN improved the determination coefficient by 132.5 % and 36.8 %, respectively. In the 48-hour APD forecasts, the VMD and TCN improved the determination coefficient by 119.7 % and 40.9 %, respectively.

Physics-informed neural networks (PINNs) are applied to solve forced-KdV (fKdV) equations and to assess their capability for nonlinear wave dynamics under external forcing. We first validate the solver in analytically tractable forced-KdV benchmarks with periodic and exponential time-dependent forcing, demonstrating accurate reproduction of two-soliton interactions. Robustness is sensitive to forcing characteristics and to the coupled choice of loss weights and sampling density: more rapidly varying forcing and imbalanced weighting increase across-seed variability in the attained relative $$\:{L}^{2}$$ error, consistent with known PINN optimization pathologies. We then demonstrate an event-based ocean application for internal solitary wave (ISW) variability near Dongsha Atoll using a reduced-order, one-dimensional fKdV representation. A spatiotemporal forcing proxy $$\:F\left(x,\:t\right)$$ is constructed from TPXO $$\:{M}_{2}$$ barotropic currents and the along-transect bathymetric slope; a ridge-based partitioning provides a consistent ridge-to-site linkage across events. A forced–free ablation of the propagation-stage forcing shows that West-Ridge corridor forcing primarily modulates site waveform morphology—systematically reducing peak magnitude while broadening the main pulse—without qualitatively altering the westward propagation pathway. These results suggest that PINNs provide a practical framework for investigating fKdV dynamics in both controlled benchmarks and mechanism-oriented ocean case studies.