ST-LSTM-SA:A New Ocean Sound Velocity Field Prediction Model Based on Deep Learning

The scarcity of in-situ ocean observations poses a challenge for real-time information acquisition in the ocean.Among the crucial hydroacoustic environmental parameters,ocean sound velocity exhibits significant spatial and temporal variability and it is highly relevant to oceanic research.In this study,we propose a new data-driven approach,leveraging deep learning techniques,for the prediction of sound velocity fields(SVFs).Our novel spatiotemporal prediction model,STLSTM-SA,combines Spatiotemporal Long Short-Term Memory(ST-LSTM) with a self-attention mechanism to enable accurate and real-time prediction of SVFs.To circumvent the limited amount of observational data,we employ transfer learning by first training the model using reanalysis datasets,followed by fine-tuning it using in-situ analysis data to obtain the final prediction model.By utilizing the historical 12-month SVFs as input,our model predicts the SVFs for the subsequent three months.We compare the performance of five models:Artificial Neural Networks(ANN),Long ShortTerm Memory(LSTM),Convolutional LSTM(ConvLSTM),ST-LSTM,and our proposed ST-LSTM-SA model in a test experiment spanning 2019 to 2022.Our results demonstrate that the ST-LSTM-SA model significantly improves the prediction accuracy and stability of sound velocity in both temporal and spatial dimensions.The ST-LSTM-SA model not only accurately predicts the ocean sound velocity field(SVF),but also provides valuable insights for spatiotemporal prediction of other oceanic environmental variables.

Rules Mining-Based Gene Expression Programming for the Multi-Skill Resource Constrained Project Scheduling Problem

Themulti-skill resource-constrained project scheduling problem(MS-RCPSP)is a significantmanagement science problem that extends from the resource-constrained project scheduling problem(RCPSP)and is integrated with a real project and production environment.To solve MS-RCPSP,it is an efficient method to use dispatching rules combined with a parallel scheduling mechanism to generate a scheduling scheme.This paper proposes an improved gene expression programming(IGEP)approach to explore newly dispatching rules that can broadly solve MS-RCPSP.A new backward traversal decoding mechanism,and several neighborhood operators are applied in IGEP.The backward traversal decoding mechanism dramatically reduces the space complexity in the decoding process,and improves the algorithm’s performance.Several neighborhood operators improve the exploration of the potential search space.The experiment takes the intelligent multi-objective project scheduling environment(iMOPSE)benchmark dataset as the training set and testing set of IGEP.Ten newly dispatching rules are discovered and extracted by IGEP,and eight out of ten are superior to other typical dispatching rules.

Multibeam water column data research in the Taixinan Basin:Implications for the potential occurrence of natural gas hydrate

A multi beam sonar survey is carried out in the continental slope of the Taixinan Basin to obtain submarine topographic and water column data. The data are processed to obtain water column images. Anomalous water column images, displaying plume characteristics, are found in gas hydrate enriched areas in the Taixinan Basin.This indicates the presence of natural gas resources in the Taixinan Basin. The multibeam sonar system is shown to provide an accurate and effective approach for detecting sub-sea gas hydrate.

Accuracy assessment of global ocean tide models in the South China Sea using satellite altimeter and tide gauge data

In this study,to meet the need for accurate tidal prediction,the accuracy of global ocean tide models was assessed in the South China Sea(0°–26°N,99°–121°E).Seven tide models,namely,DTU10,EOT11 a,FES2014,GOT4.8,HAMTIDE12,OSU12 and TPXO8,were considered.The accuracy of eight major tidal constituents(i.e.,Q1,O1,P1,K1,N2,M2,S2 and K2)were assessed for the shallow water and coastal areas based on the tidal constants derived from multi-mission satellite altimetry(TOPEX and Jason series)and tide gauge observations.The root mean square values of each constituent between satellite-derived tidal constants and tide models were found in the range of 0.72–1.90 cm in the deep ocean(depth>200 m)and 1.18–5.63 cm in shallow water area(depth<200 m).Large inter-model discrepancies were noted in the Strait of Malacca and the Taiwan Strait,which could be attributable to the complicated hydrodynamic systems and the paucity of high-quality satellite altimetry data.In coastal regions,an accuracy performance was investigated using tidal results from 37 tide gauge stations.The root sum square values were in the range of 9.35–19.11 cm,with the FES2014 model exhibiting slightly superior performance.

Detailed seafloor geomorphology of the western region of the North Yellow Sea,China:The result of Holocene erosional and depositional processes sculpting the offshore continental shelf

High-resolution multi-beam/single-beam bathymetric data and seismic profiling data from the latest surveys are used to map and interpret the detailed seafloor geomorphology of the western region of the North Yellow Sea(NYS),China.The mapping area covers 156410 km^(2),and incorporates a flat shelf plain,subaqueous accumulation shoals,tidal scouring troughs,and tidal sand ridge groups.Offshore areas with water depths less than 50 m in the western region of the NYS are mainly covered by thick,loose sediments,forming wide spread accumulation geomorphological features;these include the Liaodong Peninsula subaqueous accumulation system containing shoals and rugged scouring troughs,and the large mud wedge of the Shandong Peninsula.In the central part of the NYS,there is a relatively flat residual shelf plain with coarser sediment deposits.This flat shelf plain has a water depth larger than 50 m and a thin layer of sediment,on which there is a large pockmark field caused by seafloor seepage.These geomorphological structures indicate that modern sedimentary processes are the main driving force controlling the sculpture of the current seafloor surface landform.Extensive strong tidal current systems and abundant sediment sources provide the critical external forces and essential conditions for the formation of seafloor geomorphology.The tectonic basement controls the macroscopic morphological shape of the NYS,but is reflected very little in the seafloor geomorphic elements.Our results provide a detailed seafloor geomorphological map of the western region of the NYS,an area that has not previously mapped and also provide a scientific framework for further research into offshore seafloor geomorphology,shelf sedimentary processes,and submarine engineering construction in this region.