Multistep-Ahead Prediction of Ocean SSTA Based on Hybrid Empirical Mode Decomposition and Gated Recurrent Unit Model

Liu, Xiaoyin; Li, Ning; Guo, J.; Fan, Zhongyong; Luo, Xiaoping; Liu, Weifeng; Liu, Baodi

doi:10.1109/jstars.2022.3201228

Cited by 8 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most accurate result at lag 12, when considering all months in the evaluation period (Fig. 5), is shown to be associated with predictions from the months of May to August when they were relatively superior lead times (Liu et al 2022(Liu et al , 2023Kim et al 2022;Mu et al 2021;Chen et al 2023b;Chen et al 2023;Jonnalagadda and Hashemi 2023). Ham et al (2019) applied a convolutional LSTM to achieve skillful ENSO forecasts for lead times of up to one and a half years.…”

Section: Resultsmentioning

confidence: 98%

“…In recent years there has been a burgeoning in the application of Artificial Neural Networks (ANN) in climate science, including for ENSO prediction (Kim et al 2022;Liu et al 2022;Zhou and Zhang 2023;Jonnalagadda and Hashemi 2023). ANNs have shown potential in capturing non-linear relationships intrinsic to ENSO (Ham et al 2019;Zhao and Sun 2022;Zhang et al 2022).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning with autoencoders and LSTM for ENSO forecasting

Ibebuchi,

Richman

2024

Clim Dyn

View full text Add to dashboard Cite

El Niño Southern Oscillation (ENSO) is the prominent recurrent climatic pattern in the tropical Pacific Ocean with global impacts on regional climates. This study utilizes deep learning to predict the Niño 3.4 index by encoding non-linear sea surface temperature patterns in the tropical Pacific using an autoencoder neural network. The resulting encoded patterns identify crucial centers of action in the Pacific that serve as predictors of the ENSO mode. These patterns are utilized as predictors for forecasting the Niño 3.4 index with a lead time of at least 6 months using the Long Short-Term Memory (LSTM) deep learning model. The analysis uncovers multiple non-linear dipole patterns in the tropical Pacific, with anomalies that are both regionalized and latitudinally oriented that should support a single inter-tropical convergence zone for modeling efforts. Leveraging these encoded patterns as predictors, the LSTM - trained on monthly data from 1950 to 2007 and tested from 2008 to 2022 - shows fidelity in predicting the Niño 3.4 index. The encoded patterns captured the annual cycle of ENSO with a 0.94 correlation between the actual and predicted Niño 3.4 index for lag 12 and 0.91 for lags 6 and 18. Additionally, the 6-month lag predictions excel in detecting extreme ENSO events, achieving an 85% hit rate, outperforming the 70% hit rate at lag 12 and 55% hit rate at lag 18. The prediction accuracy peaks from November to March, with correlations ranging from 0.94 to 0.96. The average correlations in the boreal spring were as large as 0.84, indicating the method has the capability to decrease the spring predictability barrier.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

Deep learning with autoencoders and LSTM for ENSO forecasting

Ibebuchi,

Richman

2024

Clim Dyn

View full text Add to dashboard Cite

show abstract

“…The multi‐step forecasting errors are mainly caused by the overlay of rolling forecasting errors (Chen et al., 2022). To reduce the influence of rolling forecasting errors, we use the sliding forecasting method to carry out the multi‐step forecasting of the DNN models, and the forecasting accuracy of this method is less affected by the step size (e.g., X. Y. Liu et al., 2022a; Xiong et al., 2021). Table 2 shows that the input step size of GRU, LSTM, and the PRO model is 24, so the sliding step size is set at 24.…”

Section: Resultsmentioning

confidence: 99%

“…The GRU model is a simplified form of the LSTM model, with two gate structures: reset and update gates (Chung et al., 2014). Like the LSTM model, it can solve the gradient explosion problem of RNN and is also widely used in ionospheric parameter prediction (Kaselimi et al., 2022; X. Y. Liu et al., 2022a). This paper adopted the GRU model and AMTB‐2013 (Altadill et al., 2013) of the IRI‐2016 model for comparative analysis.…”

Section: Resultsmentioning

confidence: 99%

A Hybrid Deep Learning‐Based Forecasting Model for the Peak Height of Ionospheric F2 Layer

Shi,

Yang,

Wang

et al. 2023

Space Weather

View full text Add to dashboard Cite

To achieve accurate forecasting of the peak height of the ionospheric F2 layer (hmF2), we propose a hybrid deep learning model of improved seagull optimization algorithm (ISOA) optimized long short‐term memory (LSTM) model based on a complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) theory. The hybrid model decomposes the hmF2 time data into multiple subsequences through CEEMDAN and reconstructs the subsequences by sample entropy and correlation coefficient into high and low‐frequency sequences, which effectively shortens the calculation time of the model. Then, we determine the optimal hyperparameters of the LSTM models through ISOA, achieving high‐precision forecasting of the hmF2. In single‐step forecasting, the forecasting values of the hybrid model in diurnal and seasonal changes are highly consistent with the observation, which can better capture the severe changes in the hmF2. The model's RMSE, MAE, MAPE, and CC evaluation metrics are 15.86, 11.03 km, 4.76%, and 0.93 in the test set. Compared to IRI, GRU, and LSTM models, taking RMSE as an example, the forecasting accuracy of the models increased by 65.24%, 29.89%, and 29.60%, respectively. In multi‐step forecasting, the proposed model is better at forecasting the changing trend of hmF2, and the forecasting accuracies are significantly better than the IRI model. The data from multiple stations also verified the applicability of the proposed model for hmF2 forecasting. The above results indicate that the hybrid model has high accuracy in hmF2 short‐term forecasting and good applicability in multiple multi‐step forecasting, which can further improve the accurate forecasting of space weather.

show abstract

“…Jahanbakht designed an ensemble of two staked DNNs that used air temperature and SST to predict SST [27]. To forecast multi-step-ahead SST, a hybrid empirical model and gated recurrent unit was proposed [28]. Accuracy comparable to existing state of the art can be achieved by combining automated feature extraction and machine learning models [8].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Sea Surface Temperature by Combining Interdimensional and Self-Attention with Neural Networks

Guo

Wang³

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

Sea surface temperature (SST) is one of the most important and widely used physical parameters for oceanography and meteorology. To obtain SST, in addition to direct measurement, remote sensing, and numerical models, a variety of data-driven models have been developed with a wealth of SST data being accumulated. As oceans are comprehensive and complex dynamic systems, the distribution and variation of SST are affected by various factors. To overcome this challenge and improve the prediction accuracy, a multi-variable long short-term memory (LSTM) model is proposed which takes wind speed and air pressure at sea level together with SST as inputs. Furthermore, two attention mechanisms are introduced to optimize the model. An interdimensional attention strategy, which is similar to the positional encoding matrix, is utilized to focus on important historical moments of multi-dimensional input; a self-attention strategy is adopted to smooth the data during the training process. Forty-three-year monthly mean SST and meteorological data from the fifth-generation ECMWF (European Centre for Medium-Range Weather Forecasts) reanalysis (ERA5) are collected to train and test the model for the sea areas around China. The performance of the model is evaluated in terms of different statistical parameters, namely the coefficient of determination, root mean squared error, mean absolute error and mean average percentage error, with a range of 0.9138–0.991, 0.3928–0.8789, 0.3213–0.6803, and 0.1067–0.2336, respectively. The prediction results indicate that it is superior to the LSTM-only model and models taking SST only as input, and confirm that our model is promising for oceanography and meteorology investigation.

show abstract

Multistep-Ahead Prediction of Ocean SSTA Based on Hybrid Empirical Mode Decomposition and Gated Recurrent Unit Model

Cited by 8 publications

References 38 publications

Deep learning with autoencoders and LSTM for ENSO forecasting

Deep learning with autoencoders and LSTM for ENSO forecasting

A Hybrid Deep Learning‐Based Forecasting Model for the Peak Height of Ionospheric F2 Layer

Prediction of Sea Surface Temperature by Combining Interdimensional and Self-Attention with Neural Networks

Contact Info

Product

Resources

About