Lagrangian particle tracking experiments were conducted to evaluate the transport of microplastics (MPs) derived from the four major rivers, i.e., Pearl , Han (China), Mekong (Vietnam), Pasig (Philippines) rivers, that have been known to discharge a huge amount of plastic wastes into the South China Sea (SCS). We carried out two types of the experiments: 1) 2D tracking of MP particles placed at surface to represent positively buoyant (light) MPs, and 2) 3D tracking of neutrally buoyant MP particles that are passively transported by ambient flow, using the pre-computed 3D current velocity field by the HYCOM-ROMS downscaling model at a lateral resolution of 5 km for the SCS. We released more than 1 x 10⁵ particles continuously for the four-year period from 2012 to 2015 into the synoptic model and for one year into the climatological model. The MP particles are discharged at the mouths of the four major MP-source rivers located in the SCS. The comparative synoptic and climatological experiment enables us to analyze the seasonal and inter-annual variability of the river-derived MP transport. The seasonally varying monsoons in the SCS are found to provoke strong seasonal variability in the river-derived MP transport. We found that mesoscale eddies generated along the east coast of the Philippines promote westward transport and accumulation of the MPs from Pasig river. The comparative experiments also allow extracting medium-term climatological variability in the MP transport in the El Niño years (2014-2015).

The Indian Ocean (IO) is the mother of ~30% of the world’s population, who depend on marine and coastal biodiversity for their livelihoods. It is estimated that a large percentage of global plastic waste is estimated to entering the IO through different pathways. Since the ocean dynamics and air-sea interaction are unique in the IO, the dynamics of plastic debris is different in the IO than in the other global oceans. In this study, an attempt has been made to comprehend the role of near-surface currents on the transport and dispersal pattern of microplastics (MPs) in the northern Indian Ocean (NIO). Meta-data of MPs in the surface water of NIO is used to describe the dispersal pattern. We have also retrieved and used satellite remote sensing data from online databases, which consist of sea surface currents, surface winds, sea surface temperature, sea surface salinity, mean sea level anomaly, mixed layer depth and wave heights during southwest (SW), northeast (NE) and inter-monsoon (IM) seasons in the NIO. During SW monsoon season, the floating MP particles in the Arabian Sea (AS) are transported to the Bay of Bengal (BoB). The concentration of MPs is relatively higher in the BoB throughout the year because of huge volume of plastic inputs from neighboring countries, annual mean eastward flow in the equatorial region and anti-cyclonic and cyclonic gyres in the BoB which trap the plastics. During the IM season, the floating MP particles in the NIO are transported to the southern Indian Ocean (SIO) through equator along the eastern side of IO. The escape of MP particles could be due to equatorial Wyrtki jets (WJ), South Java Current (SJC) and South Equatorial current (SEC). The primary results are thought provoking and more attention is required to better understand the dispersal pattern.

The southwest East/Japan Sea (EJS) along the Korean coast is a critical fishing ground and is thought to be more productive than other regions of the EJS. Various hypotheses have been proposed to explain the region’s relatively high productivity, such as wind-driven coastal upwelling, Ulleung warm eddy, subpolar frontal dynamics, and current-induced bottom Ekman transport. Those processes can be either attributed to or modulated by Tsushima Warm Current (TWC), a northeastward-flowing branch of the Kuroshio in the EJS. To elucidate the roles of TWC on ocean primary productivity, we develop a high-resolution physical-biogeochemical coupled model using the Coastal and Regional Ocean Community model (CROCO), an evolution of the Regional Ocean Modeling System (ROMS). Model results capture the general seasonality and spatial variability in satellite-derived sea surface temperature and chlorophyll-a, as well as observed phytoplankton biomass and size composition. The physical structures (e.g., mixing, stratification, and eddy dynamics) and biogeochemical responses (nutrient supply and primary production) under varying TWC transport and temperature/salinity scenarios are examined. The relative contributions of wind- versus current-induced upwelling in the euphotic-zone nutrient budget are also quantified. The modeling analyses allow the diagnose of physical-biological processes that contribute to elevated productivity in the southwest EJS and have important implications for managing essential fisheries resources in a changing climate. 

The East China Sea is an area where red tides occur frequently. The Changjiang Diluted Water, the exchange of adjacent sea areas, the exchange of seawater and seabed sediments, and the transportation of the atmosphere are all important sources of nutrients. The atmospheric transport of trace elements in aerosols is of great significance to the primary productivity of the ocean. Phosphorus is very important to marine organisms and is also a key factor affecting red tides. This paper uses the measured data of phosphate oxygen isotope in the East China Sea to analyze the source of phosphate and study the contribution of aerosol transport to the phosphate in the East China Sea in summer. The results show that during the study period, the phosphate in the East China Sea was greatly affected by the exchange of adjacent seas and atmospheric transportation, which satisfies the mixing model. The measured δ¹⁸O_p is calculated with the mixing value calculated by the hybrid model and the formula. The comparison of the theoretical balance value shows that compared with the input of phosphorus in the surveyed sea area, the biological recycling rate of phosphorus is slow, and the phosphorus is not fully utilized, which can be used to trace the source of phosphorus. The measured value also shows that the Taiwan warm current and atmosphere Subsidence is the two main sources of phosphate in surface seawater.

As a common feature of vertical chlorophyll profiles in stratified water columns, the subsurface chlorophyll maxima (SCMs) are of great significance to the biomass and primary production. To make up for the shortcoming that remote sensing cannot detect the SCMs, an improved deep neural network (IDNN) model is used to retrieve SCMs in this study. First, the IDNN model was modulated and trained by Bio-Argo data, in which sea surface temperature (SST) and sea surface Chl a are used as input variables. The data were collected in the Northwest Pacific International Bio-Argo (120^oE-170^oE, 15^oN-45^oN), and the modeling domain was divided into three box areas in terms of latitude. The modeling results agreed well with the observations of Chl a profiles. The adaption of Gaussian activation function in the IDNN model reduced the inversion error of SCMs by 0.2. In addition, the parameter of correction bias is introduced into the IDNN model to simulate seasonal variation of SCMs well, and the results remse is within 0.2. Then the trained IDNN model is transformed into the application of remote sensing surface data in the northwest Pacific, and the result rmse is within 0.1.

Substantial studies are implemented to quantify the future global and regional mean sea level (MSL) projection. However, the discussions for the projected extreme sea level (ESL) changes are limited, which have more significant environmental and socio-economic impacts to densely populated coastal region like the coast of the China marginal seas. To investigate the projected ESL changes and how surface forcing affects the ESL, a high-resolution (~7 km) regional ocean model based on Regional Ocean Modeling System (ROMS) is set up in the China marginal seas. This regional model is forced by the fully-coupled climate model ACCESS1-0 for both the historical (1984-2005) and the future (2079-2100) periods. The validation for historical simulation against 28 tide gauge records indicates that this regional model captures well the statistical characteristics of the historical sea level distribution. Then the future sea level probability density function (PDF) changes are examined through checking the differences between historical and future experiments. We find the MSL changes explain most of the ESL changes in the future, while the PDF shape change (e.g., skewness and kurtosis) also affects the ESL in some regions. Carefully-designed perturbation experiments indicate the MSL change is mainly induced by the open boundary condition change, which contributes little to the PDF shape change. The perturbation experiments also demonstrate that the ESL is sensitive to high-frequency signal (higher than 3-month) of surface forcing, in which the wind stress changes the PDF shape significantly. The high-frequency surface forcing influences the ESL by impacting the MSL, seasonal and non-seasonal sea level signals, in which the non-seasonal signal is the largest contributor to the ESL change. We also find the high-frequency signal change in the future is a major contributor to the PDF shape change of ESL.

The Indian Ocean Dipole (IOD) significantly contributes to the interannual variability in the Indian Ocean (IO) and Indian summer monsoon. Its future changes are projected to impact the climate of the influencing region like Indian subcontinents. It is regulated and modulated by complex air-sea interaction and internal variability of IO. Understanding the IOD response on a regional scale using the regional earth system model has been largely limited. The correct representation of air-sea coupling is vital for the accurate prediction of coupled processes like IOD. In this study, a high-resolution regional earth system model (RESM), namely ROM, is employed to simulate the IOD and investigate its anomalous response on the Indian Ocean’s surface and subsurface characteristics.  Regional model, REMO, and Max-Planck Institute Model, MPIOM is taken as atmospheric and ocean components of the RESM, which integrated at 0.22⁰ for recent 38 years.  It is observed that ROM performs well in simulating the IOD events with some systematic biases in their strength. Apart from this, an attempt is also made to investigate the association of IOD with thermocline depth-subsurface temperature feedback and its relationship with El Niño-Southern Oscillation.