A model for predicting SST in limited region-I. The dynamical equations

-Starting from physical oceanology characteristics of the China seas and for the short-term operational prediction of SST in the region, a two-dimensional (vertically integrated) primitive equation model, physically reasonable and operationally feasible,on the upper mixed layer is constructed and given here, which consists of three parts, the nondivergent residual current (the monthly mean field of the Kuroshio and its branches) equations, the dynamic forecasting equations, and the equation of model's physics consisting of surface heat flux, coolings of the upper mixed layer due to the Ekman pumping and the entrainment by gale. This model may be used primarily to forecast the sea surface temperature, and to give estimations of the mean wind-driven current and the sea level, for a period of 3-5 d. In part 1 of this series, the physical conditions for establishing model equations are discussed first, that is, 1. the existence of the upper well mixed layer in the region; 2. the distinguishability of currents of all kinds; 3. the splitting of thermodynamical equation. The equations of nondivergent residual current, and the dynamic forecasting equations with initial values and boundary conditions are also discussed.

Effective ensemble learning approach for SST field prediction using attention-based PredRNN

Accurate prediction of sea surface temperature (SST) is extremely important for forecasting oceanic environmental events and for ocean studies. However, the existing SST prediction methods do not consider the seasonal periodicity and abnormal fluctuation characteristics of SST or the importance of historical SST data from different times;thus, these methods suffer from low prediction accuracy. To solve this problem, we comprehensively consider the effects of seasonal periodicity and abnormal fluctuation characteristics of SST data, as well as the influence of historical data in different periods, on prediction accuracy. We propose a novel ensemble learning approach that combines the Predictive Recurrent Neural Network(PredRNN) network and an attention mechanism for effective SST field prediction. In this approach, the XGBoost model is used to learn the long-period fluctuation law of SST and to extract seasonal periodic features from SST data. The exponential smoothing method is used to mitigate the impact of severely abnormal SST fluctuations and extract the a priori features of SST data. The outputs of the two aforementioned models and the original SST data are stacked and used as inputs for the next model, the PredRNN network. PredRNN is the most recently developed spatiotemporal deep learning network, which simulates both spatial and temporal representations and is capable of transferring memory across layers and time steps. Therefore, we used it to extract the spatiotemporal correlations of SST data and predict future SSTs. Finally, an attention mechanism is added to capture the importance of different historical SST data, weigh the output of each step of the PredRNN network, and improve the prediction accuracy. The experimental results on two ocean datasets confirm that the proposed approach achieves higher training efficiency and prediction accuracy than the existing SST field prediction approaches do.

Granulation-based LSTM-RF combination model for hourly seasurface temperature prediction

Accurate predictions of sea surface temperature(SST)are crucial due to the significant impact of SST on the global ocean-atmospheric system and its potential to trigger extreme weather events.Many existing machine-learning-based ssT predictions adapt the traditional iterative point-wise prediction mechanism,whose predicting horizons and accuracy are limited owing to the high sensitivity to cumulative errors during iterative predictions.Therefore,this paper proposes a novel granulation-based long short-term memory(LsTM)-random forest(RF)combination model that can fully capture the feature dependencies involved in the fluctuation of SsT sequences,reduce the cumulative error in the iteration process,and extend the prediction horizons,which includes two sub-models(adaptive granulation model and hybrid prediction model).They can restack the one-dimensional ssT time-series into multidimensional feature variables,and achieve a strong forecasting ability.The analysis shows that the proposed model can achieve more accurate prediction-hours in nearly all prediction ranges from 1 to 125 h.The average prediction error of the proposed model in 25-125 h is 0.07 K,similar to that(0.067 K)in the first 24 h,which exhibits a high generalization performance and robustness and isthus a promising platform for the medium-and long-term forecasting of hourly SSTs.