The exchange flow structure was examined in the North Passage of Changjiang River Estuary, where a deep waterway project (DWP) was carried out to improve the navigability. Before the construction of the DWP, the friction effect played a significant role in shaping the transverse structure of the exchange flow. The turbulent eddy viscosity generated near the seabed can be transferred to the upper water column, which facilitated vertical momentum exchange. As a result, the landward inflow extended to –2 m below the water surface and the seaward outflow was concentrated on the shallow shoal on the southern side of the cross section. After the construction of the DWP, the turbulent mixing was suppressed as a result of density stratification. The friction felt by the water was constrained in the lower half of the water column and the vertical momentum exchange was reduced. Meanwhile, the channel became dynamically narrowed with a Kelvin number of 0.52. Therefore, the Coriolis played a minor role in shaping the transverse structure of the exchange flow. As a consequence, the exchange flow featured a vertically-sheared pattern, with outflow at the surface and inflow underneath. Additionally, the gravitational circulation was enhanced due to increase in along-channel density gradient and stratification. The exchange flow components associated with the lateral processes (residual currents induced by eddy viscosity-shear covariance and lateral advective acceleration) were reduced, which suggests that lateral processes played a minor role in modifying the along-channel dynamics when the estuary becomes dynamically-narrowed.

Conventional retrieval and neural network methods are used simultaneously to retrieve sea surface wind speed (SSWS) from HH-polarized Sentinel-1 (S1) SAR images. The Polarization Ratio (PR) models combined with the CMOD5.N Geophysical Model Function (GMF) is used for SSWS retrieval from the HH-polarized SAR data. We compared different PR models developed based on previous C-band SAR data in HH-polarization for their applications to the S1 SAR data. The recently proposed CMODH, i.e., retrieving SSWS directly from the HH-polarized S1 data is also validated. The results indicate that the CMODH model performs better than results achieved using the PR models. We proposed a neural network method based on the backward propagation (BP) neural network to retrieve SSWS from the S1 HH-polarized data. The SSWS retrieved using the BP neural network model agrees better with the buoy measurements and ASCAT dataset than the results achieved using the conventional methods. Compared to the buoy measurements, the bias, root mean square error (RMSE) and scatter index (SI) of wind speed retrieved by the BP neural network model are 0.10 m/s, 1.38 m/s and 19.85%, respectively, while compared to the ASCAT dataset the three parameters of training set are –0.01 m/s, 1.33 m/s and 15.10%, respectively. It is suggested that the BP neural network model has a potential application in retrieving SSWS from Sentinel-1 images acquired at HH-polarization.

The evolution of thermohaline structure at the upper ocean during three tropical cyclones (TCs) in the Northwest Pacific was studied in this study based on successive observation by two new-style underwater gliders during fall 2018. These remote-controllable gliders with CTD sensor enabled us to explore high frequency responses of temperature, salinity, mixed and barrier layers in the upper ocean to severe TCs in this area. Results showed that three significant cooling-to-warming and stratification destructing-to-reconstructing processes at the mixed layer occurred during the lives of three TCs. The maximal cooling of SST all reached ≥0.5°C although TCs with different intensities had different minimal distances to the observed area. Under potential impacts of solar radiation, tide and inertial motions, the mixed layer depth possessed significant high-frequency fluctuations during TC periods. In addition, barrier layers appeared and vanished quickly during TCs, accompanied with varied temperature inversion processes.

Two kinds of regression equations are used to reproduce the sediment flux of the 26 small coastal watersheds in southeastern China. The first kind is the global equations suggested by Milliman and Syvitski (1992), Mulder and Syvitski (1996), Syvitski et al. (2003), and Syvitski and Milliman (2007). The second kind is the modified equations revised by the characteristics of the coastal watersheds, including the drainage area, mean water discharge, and mean sediment discharge. Compared with the observations of the hydrometric stations, the global equations overestimate the sediment flux by 1–2 orders of magnitude. By using the modified equations, the accuracy of the estimated sediment flux is significantly improved, with the relative error in the range of 7%–24%. The reason for the overestimation mainly caused by different parameters’ domain and regression coefficients between global rivers and study coastal watersheds. This study demonstrates that modification needs to be considered when using global regression equations to reproduce the sediment flux of the small coastal watersheds in southeastern China.

Based on the latest oceanic surface drifter dataset from the global drifter program during 2000–2019, this study investigated the global variation of relative frequency shift (RFS), near-inertial energy (NIE) and inverse excess bandwidth (IEB) of near-inertial motions, and analyzed their relations with oceanic mesoscale dynamics, relative vorticity and strain. Compared with previous works, we have some new findings in this study: (1) the RFS was high with negative values in some regions in which we found a significant blue shift of the RFS in the equatorward of 30°N (S) and from 50°N to 60°N in the Pacific, and a red shift in the western boundary currents and their extension regions, the North Atlantic and the Antarctic Circumpolar Current regions; (2) more peak values of the NIE were found in global regions like the South Indian Ocean, the Luzon Strait and some areas of the South Ocean; (3) the global distribution of the IEB were characterized by clear zonal bands and affected by vorticity and wind field; (4) the RFS was elevated as the absolute value of the gradient of vorticity increased, the IEB did not depend on the gradient of vorticity, and the eddy kinetic energy (EKE) weakened with the decrease of the absolute value of RFS; (5) the NIE decreased with increasing absolute value of the relative vorticity and the gradient of vorticity, but it increased with increasing strain and EKE when EKE was larger than 0.003 2 m²/s².

To explore new operational forecasting methods of waves, a forecasting model for wave heights at three stations in the Bohai Sea has been developed. This model is based on long short-term memory (LSTM) neural network with sea surface wind and wave heights as training samples. The prediction performance of the model is evaluated, and the error analysis shows that when using the same set of numerically predicted sea surface wind as input, the prediction error produced by the proposed LSTM model at Sta. N01 is 20%, 18% and 23% lower than the conventional numerical wave models in terms of the total root mean square error (RMSE), scatter index (SI) and mean absolute error (MAE), respectively. Particularly, for significant wave height in the range of 3–5 m, the prediction accuracy of the LSTM model is improved the most remarkably, with RMSE, SI and MAE all decreasing by 24%. It is also evident that the numbers of hidden neurons, the numbers of buoys used and the time length of training samples all have impact on the prediction accuracy. However, the prediction does not necessary improve with the increase of number of hidden neurons or number of buoys used. The experiment trained by data with the longest time length is found to perform the best overall compared to other experiments with a shorter time length for training. Overall, long short-term memory neural network was proved to be a very promising method for future development and applications in wave forecasting.

A comprehensive analysis of sea ice and its snow cover during the summer in the Arctic Pacific sector was conducted using the observations recorded during the 7th Chinese National Arctic Research Expedition (CHIANRE-2016) and the satellite-derived parameters of the melt pond fraction (MPF) and snow grain size (SGS) from MODIS data. The results show that there were many low-concentration ice areas in the south of 78°N, while the ice concentration and thickness increased significantly with the latitude above the north of 78°N during CHIANRE-2016. The average MPF presented a trend of increasing in June and then decreasing in early September for 2016. The average snow depth on sea ice increased with latitude in the Arctic Pacific sector. We found a widely developed depth hoar layer in the snow stratigraphic profiles. The average SGS generally increased from June to early August and then decreased from August to September in 2016, and two valley values appeared during this period due to snowfall incidents.

Using a gridded array for real-time geostrophic oceanography (Argo) program float dataset, the features of upper-ocean salinity stratification in the tropical Pacific Ocean are studied. The salinity component of the squared Brunt-Väisälä frequency $ {N}^{2} $ ($ {N}_{S}^{2} $) is used to represent salinity stratification. Layer-max $ {N}_{S}^{2} $ (LMN), defined as the $ {N}_{S}^{2} $ maximum over the upper 300 m depth, and halocline depth (HD), defined as the depth where the $ {N}_{S}^{2} $ maximum is located, are used to specifically describe the intensity of salinity stratification. Salinity stratification in the Topical Pacific Ocean has both spatial and temporal variability. Over the western and eastern equatorial Pacific, the LMN has a large magnitude with a shallow HD, and both have completely opposite distributions outside of the equatorial region. An obvious seasonal cycle in the LMN occurs in the north side of eastern equatorial Pacific and freshwater flux forcing dominates the seasonal variations, followed by subsurface forcing. At the eastern edge of the western Pacific warm pool around the dateline, significant interannual variation of salinity stratification occurs and is closely related to the El Niño Southern Oscillation event. When an El Niño event occurs, the precipitation anomaly freshens sea surface and the thermocline shoaling induced by the westerly wind anomaly lifts salty water upward, together contribute to the positive salinity stratification anomaly over the eastern edge of the warm pool. The interannual variations in ocean stratification can slightly affect the propagation of first baroclinic gravity waves.

This study explores the ice flow acceleration (21.1%) of Pedersenbreen during 2016–2017 after the extremely warm winter throughout the whole Arctic in 2015/2016 using in situ data and quantitatively analyses the factors contributing to this acceleration. Several data sets, including 2008–2018 air temperature data from Ny-Ålesund, ten-year in situ GPS measurements and Elmer/Ice ice flow modelling under different ice temperature scenarios, suggest that the following factors contributed to the ice flow acceleration: the softened glacier ice caused by an increase in the air temperature (1.5°C) contributed 2.7%–30.5%, while basal lubrication contributed 69.5%–97.3%. The enhanced basal sliding was mostly due to the increased surface meltwater penetrating to the bedrock under the rising air temperature conditions; consequently, the glacier ice flow acceleration was caused mainly by an increase in subglacial water. For Pedersenbreen, there was an approximately one-year time lag between the change in air temperature and the change in glacier ice flow velocity.

Site U1446 (19°50’N, 85°44’E, at water depth 1 430 m) was drilled during Expedition 353 (Indian monsoon rainfall) of the International Ocean Discovery Program (IODP). It is located in the Mahanadi offshore basin, on the northern Bay of Bengal. Sedimentation rates and contents of biocarbonates are high at this relatively shallow site. Using a micropaleontological approach, we examined planktonic and benthic foraminifera in the upper around 40 m of this site, spanning the last around 190 ka. A striking feature of the foraminiferal record is the occurrence of strong but varying dissolution although the site is located well above the modern lysocline. Such strong dissolution has never been reported in this area. We estimated the flux of foraminifera and quantified the ratio of benthic foraminifera over total foraminifera (benthic/total foraminifera) along with the foraminifer fragmentation index in order to characterize past changes in this above-lysocline dissolution. This study reveals a clear glacial-interglacial contrast, with a stronger dissolution during marine isotope stages (MISs) 1 and 5 than during MISs 2–4 and 6. Such a difference in preservation is likely to have a strong impact on geochemical proxies measured on foraminifera. Our new observations call for an in-depth study of the causes of such above-lysocline dissolution in the region, and an evaluation of its impact on the foraminifera-based proxies used for paleoenvironmental reconstruction.