Abstract:Accurate prediction of estuarine salinity can effectively mitigate the adverse effects of saltwater intrusion and help ensure the safety of water resources in estuarine regions. Presently, diverse data-driven models, mainly neural network models, have been employed to predict tidal estuarine salinity and obtained considerable achievements. Due to the nonlinear and nonstationary features of estuarine salinity sequences, this paper proposed a multi-factor salinity prediction model using an enhanced Long Short-Term Memory (LSTM) network. To improve prediction accuracy, input variables of the model were determined through Grey Relational Analysis (GRA) combined with estuarine dynamic analysis, and hyperparameters for the LSTM model were optimized using a multi-strategy Improved Sparrow Search Algorithm (ISSA). The proposed ISSA-LSTM model was applied to predict salinity at the Cangqian and Qibao stations in the Qiantang Estuary of China, based on measured data from 2011–2012. The model performance is evaluated by mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), and Nash-Sutcliffe efficiency (NSE). The results show that compared to other models including Back Propagation neural network (BP), Gate Recurrent Unit (GRU), and LSTM model, the new model has smaller errors and higher prediction accuracy, with NSE improved by 8–32% and other metrics (MAP, MAPE, RMSE) improved by 15–67%. Meanwhile, compared with LSTM optimized with the original SSA (SSA-LSTM), MAE, MAPE, and RMSE values of the new model decreased by 13–16%, 15–16%, and 11–13%, and NSE value increased by 5–6%, indicating that the ISSA has a better hyperparameter optimization ability than the original SSA. Thus, the model provides a practical solution for the rapid and precise prediction of estuarine salinity.

Prediction of Salinity in Qiantang Estuary Based on Wavelet Neural Network Optimized by Particle Swarm Optimization

A Hybrid Methodology for Salinity Time Series Forecasting Based on Wavelet Transform and Narx Neural Networks

Salinity Prediction Based on Improved LSTM Model in the Qiantang Estuary, China

Salinity Time Series Prediction and Forecasting Using Dynamic Neural Networks in the Qiantang River Estuary

Wavelet Neural Network for Modeling Chlorophyll a Concentration Affected by Artificial Upwelling

Research on Short-Term Prediction of Typical Trial Sea Environment in China Based on LSTM Neural Network

Prediction of Water Quality Index in Qiantang River Based on BP Neural Network Model

Predicting the impact of sea level rise on saltwater intrusion in the Qiantang Estuary

Neural network modeling of salinity in Qiantang Estuary

Prediction of River Sediment Transport Based on Wavelet Transform and Neural Network Model

Attribution analysis and forecast of salinity intrusion in the Modaomen estuary of the Pearl River Delta

Prediction model of aquaculture water temperature and pH based on BP neural network optimized by particle swarm algorithm

Multistep-ahead Forecasting of Chlorophyll a Using a Wavelet Nonlinear Autoregressive Network

Optimized BP Neural Network for Dissolved Oxygen Prediction

Improved BP neural network prediction model based on particle swarm algorithm

Sea-water-level Prediction Via Combined Wavelet Decomposition, Neuro-Fuzzy and Neural Networks Using SLA and Wind Information

Application of a Hybrid Optimized BP Network Model to Estimate Water Quality Parameters of Beihai Lake in Beijing

Prediction of Estuarine Water Quality Using Interpretable Machine Learning Approach

Application of Neural Network to Prediction of Coastal Water Quality

Short-Term Tidal Level Forecasting Based On Self-Adapting Pso-Bp Neural Network Model

A wavelet - Particle swarm optimization - Extreme learning machine hybrid modeling for significant wave height prediction