Multiple environmental variables related to ocean currents, freshwater runoff, and upwelling in a coastal area have complex effects on the phytoplankton community. To assess the influence of environmental variables on the phytoplankton community structure during the summer of 2019, we investigated the various abiotic and biotic factors in Korean coastal waters (KCWs), separated into five different zones. Summer environmental factors in KCWs were strongly influenced by Changjiang Diluted Water (CDW) in St. SO (Southern Offshore) 1 and 2, upwelling in St. SI (Southern Inshore) 2–4, and Nakdong River discharge in St. SI 12. In particular, low–salinity water masses (p < 0.05 for nearby locations) of CDW gradually expanded from the East China Sea to southwestern KCWs from June to July. In addition, there were high levels of nutrients following freshwater runoff from the Nakdong River in southeastern KCW, which led to the dominance of Cryptomonas spp. (81%), a freshwater and brackish water algae. On the other hand, upwelling areas in southwestern KCW were dominated by diatoms Skeletonema spp., and are characterized by high phosphate concentrations (p < 0.05) and low temperatures (p < 0.05) compared to nearby locations. Leptocylindrus danicus (20%) was dominant due to the effect of water temperature in the SE (Southeastern area) zone. Low nutrient concentrations were maintained in the East Sea (dissolved inorganic nitrogen (DIN) = 0.39 ± 0.40 μM; dissolved inorganic phosphate (DIP) = 0.09 ± 0.03 μM) and the Yellow Sea (DIN = 0.40 ± 0.07 μM; DIP = 0.04 ± 0.02 μM), which were characterized by low levels of chlorophyll a and dominated by unidentified small flagellates (35, 40%). Therefore, our results indicated that hydro–oceanographic events such as upwelling and freshwater run–off, but not ocean currents, provide nutrients to the euphotic layers of the coastal environment and play important roles in determining the phytoplankton community structure during summer in the KCWs.

The East Sea Intermediate Water (ESIW) is the subsurface salinity minimum layer in the depth range 200–400 m in the East/Japan Sea. The origin, variability, and pathways of the ESIW were investigated using a 4‐year‐long (2015–2018) high‐resolution (1/24°) ocean reanalysis and Lagrangian particle modeling. A backward particle tracking analysis revealed that the ESIW was formed south of Vladivostok and shows strong interannual variability. In 2018, owing to strong northwesterly winds, heat loss to the air intensified, and freshwater transport from the northern East/Japan Sea was simultaneously decreased by about 18% compared with the previous year. Furthermore, there was increased formation of water masses heavier than that of the typical ESIW, and the volume of the ESIW was reduced. A forward particle tracking analysis showed that the contribution of the particles originated from the northern end of the East/Japan Sea was about two times greater than those from the southern end. The major portion of the ESIW's freshwater came from the northern East/Japan Sea, where sea ice melted and the Amur River discharged. The remaining portion was from the Changjiang River, which came through the Korea Strait. The ESIW reached the Ulleung Basin over the course of a year mainly through the deep channel, the Ulleung Trough, located to the north of the basin. The diapycnal diffusivity of 1.0 × 10⁻⁴ m² s⁻¹ could erode away the low salinity core in about 27 years.

The prediction of sea surface currents is essential for various marine activities such as disaster monitoring, fishing industries, and search and rescue operations. Continuous improvements in numerical models have made it possible to predict more realistic oceans using data assimilation and fine spatial resolution. However, such complicated ocean models require high computational power and time, making them difficult to use for near‐real time forecasting. To compensate for these high computational costs, novel approaches with efficient computational costs combined with numerical model outputs need to be developed. Artificial neural networks could be one of the solutions as they require low computational power for prediction, owing to pre‐trained networks. Here, we present a prediction framework applicable to surface current prediction in the seas around the Korean Peninsula using three‐dimensional (3‐D) convolutional neural networks. The network is based on a 3‐D U‐shaped network structure and is modified to predict sea surface currents using oceanic and atmospheric variables. The forecast procedure is optimized to minimize the error of the next day's sea surface current field with a spatial resolution of 1/24° and its recursively predicting structure allows more days to be predicted. The network performance is evaluated by changing the input days and variables to find the optimal surface‐current‐prediction artificial neural network model. The optimized model shows a root mean squared error of 2.3 cm/s for the first forecast day, demonstrating its strong potential for practical use in the near future.