Based on National Oceanic and Atmospheric Administration daily optimum interpolation sea surface temperature (SST) data set from 1982 to 2019, spatial-temporal changes in marine heat wave (MHW) (frequency, duration, intensity) in the Bohai Sea and Yellow Sea and associated circulations were analyzed using various statistical methods. Results show that: (1) there are obvious regional differences in MHW. More frequency, longer duration, more intensity MHWs are all located at the most part of the Bohai Sea and northern Yellow Sea. (2) In recent 38 years, the annual frequency, annual average duration, annual average intensity and maximum intensity of the marine heat waves in the Bohai Sea and Yellow Sea have generally increased and strengthened, together with spatial heterogeneity. The trends at the coast of the Korean Peninsula coastal show insignificant trend. (3) MHWs were classified into four levels by daily SST: moderate, strong, severe and extreme. Result shows that except extreme MHWs, there are significant geographic differences in the frequency and trend of the other three MHW levels. Moderate MHWs are more frequent and have a significant increasing trend in most part of the Bohai Sea and Yellow Sea, while strong and severe marine heat waves are mainly concentrated in the Bohai Sea. However, extreme MHW in this region was hardly occurred. (4) Regional, there were 83 MHWs occurred in the Bohai Sea and Yellow Sea during 1982 to 2019, with an average of 2.2 times per year. The frequencies of MHWs appeared seasonal variation obviously. More MHWs with different levels of intensity occurred in summer. (5) Composite analysis shows that, summer MHWs in the Bohai Sea and Yellow Sea have significant relationship with atmospheric circulation. Barotropic positive geopotential height anomalies centered on Lake Baikal were favorable for anomalous descending air motion in the lower troposphere and west-northerly winds in high troposphere, there should be more clear days and hence more MHWs during the high-solar-radiation summer period.

The characteristics of the maximum shear force in the boundary layer were studied based on the wave attenuation experiment of nearshore plants considering the influence of roots. The maximum shear force characteristics of the boundary layer under two plant models, root alone and combination of roots and stems, were investigated by using the formula of boundary layer velocity calculation with plant water flow under regular waves proposed by predecessors. The shear force variation along the path and the effect of plants on the shear force attenuation characteristics of the two models were analyzed, and the formulas of the attenuation coefficients were fitted. Studies have shown that when waves pass through the plant zone, the shear force increases to a certain extent and then decreases gradually. The reduction effect of plant on shear force increases with the increase of incident wave height, and the stem contributes to the reduction effect of root group on the maximum shear force of boundary layer, and the reduction coefficient ranges from 0.06 to 0.61. The fitted formula is more suitable for shallow water.

Based on the adaptive boosting algorithm (AdaBoost) combined with the extreme learning machine (ELM), the strong classifier of ELM-AdaBoost with strong robustness and high precision is thus constructed by iterating, adjusting, and optimizing the weights between each ELM classifier. ELM-AdaBoost method can enhance the stability of the existing ELM classifier. In this paper, the data collected by side scan sonar in the Zhujiang River Estuary was used to classify and identify three types of typical sediments as rock, sand, and mud. The average classification accuracy of new method exceeds 90%, which is better than the average classification accuracy of a single ELM classifier of 85.95%. It is also superior to other traditional classifiers (i.e. LVQ and BP) and it takes much less time to classify than traditional classifiers. The experimental result shows that the proposed ELM-AdaBoost method can be effectively applied to the classification and identification of seabed sediment and can meet the needs of real-time classification of seabed sediment.

When performing work, underwater platforms usually need about 1-month to 3-months ocean numerical forecasts results. However, the current numerical forecasting technology is limited to the corresponding weather-driven field forecasting, and it is difficult to provide numerical forecast products for more than 10 days. Considering that the seawater itself has large inertia physically, and the inner ocean area has its own evolutionary principles, a statistical prediction method based on evolutionary operator is proposed. The method uses historical satellite remote sensing data to construct a long-term evolution matrix of ocean state variables, and then combines with the inertial prediction model. The final medium- and long-term statistical prediction model of the South China Sea is constructed, which can provide the prediction results for 1 d to 60 d of daily sea surface height anomaly in the South China Sea. We carry out the numerical experiments to show the effectiveness of the method. The results show that the correlation coefficients between the forecast results and the satellite data are all higher than 0.8 within 15 days and higher than 0.6 within 60 days after the start date of forecast.

The vertical distribution of rip current formed by the cross wave field near a groin is studied through the physical model experiment with regular and irregular waves and the characteristics of the distribution for transverse, longitudinal velocities and velocity vector at each measurement section are examined. The measurement is along the center line of the depth-average velocity profile in the rip channel. The influence of the position of the node and antinode point relative to the channel centerline is considered. The results of time-average and time-dependent values of rip curent velocity are analysed, and it is found that the longitudinal velocity has stable mean values for different lengths of time spans considered, but this is not the case for the transverse velocity. Due to the effect of rip current instablity, the latter has different mean values for different lengths time spans. This leads to different three-dimensional distribution characteristics of velocity vector over different time spans. However, there are some common points among the results of the different time spans, this includes, most of the velocity vectors at measurement sections deflect towards the groin, which reflects the influence of the groin; the vertical distribution of longitudinal velocity at different measurement sections all follow the power law.

The effects of three kinds of modulation (the tilt modulation, the hydrodynamic modulation and the velocity bunching modulation) on sea wave SAR (synthetic aperture radar) image are compared and analyzed firstly. The results show that the velocity bunching modulation has the most significant effect on SAR images. In addition, due to the inherent speckle noise in the SAR image, it is difficult to filter or suppress the noise in the range of low wavenumber. And the inversion of wave spectrum by classical MPI (max-planck institute) method will result in the larger spectral value in the range of low wavenumber. Referring to the classical MPI inversion algorithm, an azimuth slope spectrum and SWH (significant wave height) inversion algorithm based on the velocity bunching modulation are proposed in this paper. Comparing the SWH obtained by the classical MPI method, the co-polarization modulation method and the algorithm proposed in this paper with the buoy data, one can find that the mean square error between the SWH retrieved by the algorithm proposed in this paper and the SWH obtained from buoy data is 0.79 m, which is the smallest among the three methods.

Black-hole eddy can be extracted based on elliptic Lagrangian Coherent Structures (eLCSs), which can transport material and maintain coherent under geostrophic for a long time. It was similar to the black hole in the ocean, so it was called black-hole eddy. In this paper, the boundary of black-hole eddy was extracted based on the data of the geostrophic flow velocity field. Using the method of eLCSs and choosing a targeted eddy (Eddy A) in the western Pacific to analyze. Sea surface temperature, sea surface salinity and chlorophyll concentration data are used to verify that the Eddy A is coherent in horizontal material transport. The temperature, salinity and dissolved oxygen data obtained by Argo in different depths are used to prove the coherence in vertical of Eddy A. The result demonstrates that the boundary of black-hole eddy is more coherent than Eulerian boundary in a long time scale and can describe the transport of material more objectively.

The global climate and environment have changed significantly upon the development of the Northern Hemisphere glaciation (NHG) during the Pliocene/Pleistocene. The Pacific meridional overturning circulation (PMOC) plays an important role on the distribution of heat in the global ocean and atmospheric CO₂ sequestration in the deep ocean, however, the relationship between PMOC and the formation of NHG is poorly studied. In this paper, we collected the available Nd isotope data of seamount Fe-Mn crusts from the Pacific. By comparing the Nd isotopic records from different water depths and different regions of the Pacific, considering the influences of water mass evolution and dust input on the Nd isotope records, we discuss the evolution of the PMOC and its relationship with the global climate change. It is suggested that the stagnation of the deep water formation in the North Pacific and the increase of the Asian dust input may be the reasons for the decline of the εNd of the North Pacific deep water since 3～4 Ma, and the increase of CO₂ sequestration in the deep ocean caused by the weakening of the deep water ventilation in the North Pacific contribute to the global cooling and the formation of NHG.

Based on the 141 surface sediments collected from the Yazhou Bay, Hainan Island, the distribution characteristics of grain size components, grain size parameters and sediment types were analyzed. The results indicate that the samples can be classified into 6 types as argillaceous arenaceous gravel, gravelly argillaceous sand, sand, silty sand, arenaceous silt and silt in the study area. In particular, the arenaceous silt is most widely distributed. The grain size diameters increase and then decrease, showing a banding distribution from north to south. The separation coefficient is generally high, which indicates the complex sediment sources and hydrodynamic conditions in this area. Base on Flemming ternary diagram, combined with sediment sources, hydrodynamic conditions and topographic features, the study area is divided into three sedimentary districts from north to south: the nearshore zone which showing the characteristics of wave-controlled deposition is mostly impacted by wave winnowing and transport, and the grain size is relatively small; the central Yazhou Bay zone is a high-energy coarse grained sand area, mainly controlled by the combined action of runoff, wave and tidal current; relatively, the southern zone is mainly affected by tidal current load, where the wave action is weakened, the sediment is the finest and deposited in a lower energy environment. In addition, the analysis of sediment transport trend shows that the midwest part of the study area is a depositional center, the silt of coastal erosion, Ningyuan River and offshore tidal current transport are carried to this area.

Interannual variability of meridional heat transport anomaly (MHTA) in the tropical Indian Ocean (IO) during monsoon transitions (boreal spring and autumn) is investigated based on SODA3.4.2 and POP2 results. There exist two leading interannual modes in spring, namely the consistent mode and the divergence and convergence (div/conv) mode. The consistent mode characterized by the northward MHTA in the upper IO, is controlled by the antisymmetric wind structures with northeasterlies anomalous northeasterlies (northwesterlies) to north (south) of the equator. However, the spring div/conv mode is featured by the meridional divergence in the surface layer and convergence in the subsurface layer on both side of the equator. Such a mode is modulated by the sea surface temperature anomaly (SSTA) dipole in the tropical southwestern IO and subtropical southeastern IO south of the equator. In autumn, the div/conv mode is dominated by the IOD-induced equatorial easterlies and associated anticyclonic wind stress anomalies in the off-equatorial regions. The POP2 model sensitivity experiments confirm the roles of anomalous winds induced by the Indian Ocean SSTA modes in modulating the MHTA, that is, the antisymmetric winds induce northward MHTA, while the anomalous easterly winds lead to the div/conv pattern. Therefore, the interannual variability of air-sea coupled modes in the IO plays a vital role in the basin-scale and hemispheric-scale heat exchanges.