Short-term sea surface temperature (SST) forecasts are crucial for operational oceanology. This study introduces a specialized Transformer model (U-Transformer) to forecast global short-term SST variability and compares its performance with Convolutional Long Short-Term Memory (ConvLSTM) and Residual Neural Network (ResNet) models. The U-Transformer model forecast consistently outperformed the ConvLSTM and ResNet models, especially in regions with active mesoscale eddies. Globally, the U-Transformer model achieved SST root mean square errors (RMSEs) ranging from 0.2 °C at a 1-day lead time to 0.54 °C at a 10-day lead time during 2020–2022, with anomaly correlation coefficients (ACCs) decreasing from 0.97 to 0.79, respectively. However, in regions characterized by active mesoscale eddies, RMSEs from the U-Transformer model exceeded the global averages by at least 40%, with values in the Gulf Stream region reaching more than twice the global average. Additionally, ACC values in active mesoscale eddy regions declined more sharply with forecast lead time compared to the global averages, decreasing from approximately 0.96 at a 1-day lead time to 0.73 at a 10-day lead time. Specifically, the ACC value dropped to 0.89 in the Gulf Stream region at a 3-day lead time, while maintaining 0.92 globally. These findings underscore the importance of advanced approaches to enhance SST forecast accuracy in challenging active mesoscale eddy regions.

The Pacific Decadal Oscillation (PDO), as the dominant mode of decadal sea surface temperature variability in the North Pacific, exhibits both interannual and decadal fluctuations that significantly influence global climate. The complexity associated with PDO changes poses challenges for accurate predictions. This study develops a BiLSTM-WOA-MMD (BWM) model, which integrates a bidirectional long short-term memory network with a whale optimization algorithm (WOA) and multiple modal decomposition (MMD), to forecast PDO at both interannual and decadal time scales. The model successfully predicts monthly/annual average PDO index of up to 15 months/5 years in advance, achieving a correlation coefficient of 0.56/0.55. By utilizing the WOA to effectively optimize hyperparameters, the model enhances the PDO prediction skill compared to existing deep learning PDO prediction models, improving the correlation coefficient from 0.47 to 0.68 at a 6-month lead time. The combination of MMD and WOA further minimizes prediction errors and extends the forecasting effective time to 15 months by capturing essential modes. The BWM model can be employed for future PDO prediction and the predicted PDO will remain in its cool phase in the next year both using the PDO index from NECI and derived from near-time satellite data. This proposed model offers an effective way to advance the prediction skill of climate variability on multiple time scales by utilizing all kinds of data available including satellite data, and provides a large-scale background to monitor marine heatwaves.

The diurnal sea surface temperature variation (DSV) influences atmospheric convection and precipitation through air–sea interactions in the equatorial Pacific. Deep learning-based DSV forecasting has been less explored compared to traditional methods, presenting the potential for a substantial leap in forecast accuracy. In this study, a forecast model is developed for 24 h DSV in the equatorial Pacific using an improved coupled Transformer-CNN (CoTCN-DSV) by incorporating a new loss function including maximal and minimal values. The CoTCN-DSV forecasts diurnal variation in SST at the interval of 3 h based on 3 h SST from the WHOI dataset. The CoTCN-DSV captures DSV well with root mean square error (RMSE) of DSA below 0.03 °C/0.13 °C at 3 h/12 h lead times and maintains high forecast skill with the temporal correlation coefficient (R) of 0.78 at the lead times of 12 h in the equatorial Pacific. The CoTCN-DSV reduces RMSE for daily max/min SST by 10.9% and 12.8% due to replacing the new loss function, then significantly improving DSV forecast. There are systematic SST biases in the WHOI dataset and this leads to relatively large RMSEs when DSV forecasts trained using WHOI are evaluated against TAO observations. Replaced WHOI SST by TAO SST, the forecasted DSA RMSE by CoTCN-DSV is reduced by an average of 43%. This confirms that the CoTCN-DSV has good generalization ability and high-quality data are important to advance the forecast accuracy. These finding show that CoTCN-DSV has the potential to forecast extreme values for different scenarios.