In this paper, we propose a novel method to enhance the accuracy of a real-time ocean forecasting system. The proposed system consists of a real-time restoration system of satellite ocean temperature based on a deep generative inpainting network (GIN) and assimilation of satellite data with the initial fields of the numerical ocean model. The deep learning real-time ocean forecasting system is as fast as conventional forecasting systems, while also showing enhanced performance. Our results showed that the difference in temperature between in situ observation and actual forecasting results was improved by about 0.5 °C in daily average values in the open sea, which suggests that cutting back the temporal gaps between data assimilation and forecasting enhances the accuracy of the forecasting system in the open ocean. The proposed approach can provide more accurate forecasts with an efficient operation time.

In this study, the occurrence of Cochlodinium polykrikoides bloom was predicted based on spatial information. The South Sea of Korea (SSK), where C. polykrikoides bloom occurs every year, was divided into three concentrated areas. For each domain, the optimal model configuration was determined by designing a verification experiment with 1–3 convolutional neural network (CNN) layers and 50–300 training times. Finally, we predicted the occurrence of C. polykrikoides bloom based on 3 CNN layers and 300 training times that showed the best results. The experimental results for the three areas showed that the average pixel accuracy was 96.22%, mean accuracy was 91.55%, mean IU was 81.5%, and frequency weighted IU was 84.57%, all of which showed above 80% prediction accuracy, indicating the achievement of appropriate performance. Our results show that the occurrence of C. polykrikoides bloom can be derived from atmosphere and ocean forecast information.

The performance of three turbulence closure schemes (TCSs), the generic length scale scheme (GLS), the Mellor–Yamada 2.5 scheme (MY2.5) and the K-profile parameterization scheme (KPP), embedded in the ocean model ROMS, was compared with attention to the reproduction of summertime temperature distribution in the Yellow Sea. The ROMS model has a horizontal resolution of 1/30° and 30 vertical sigma layers. For model validation, root mean square errors were checked, comparing model results with wave and temperature buoy data as well as tidal station data supplied by various organizations within the Republic of Korea. Computed temperature and vertical temperature diffusion coefficients were mainly compared along Lines A (36° N) and B (125° E) crossing the central Yellow Sea, Lines C (32° N) and E (34° N) passing over the Yangtze Bank and Line D off the Taean Peninsula. Calculations showed that GLS and MY2.5 produced vertical mixing stronger than KPP in both the surface and bottom layers, but the overall results were reasonably close to each other. The lack of observational data was a hindrance in comparing the detailed performance between the TCSs. However, it was noted that the simulation capability of cold patches in the tidal mixing front can be useful in identifying the better performing turbulence closure scheme. GLS and MY2.5 clearly produced the cold patch located near the western end of Line E (122° E–122.3° E), while KPP hardly produced its presence. Similar results were obtained along Line D but with a less pronounced tidal mixing front. Along Line C, GLS and MY2.5 produced a cold patch on the western slope of the Yellow Sea, the presence of which had never been reported. Additional measurements near 125° E–126° E of Line C and along the channel off the Taean Peninsula (Line D) are recommended to ensure the relative performance superiority between the TCSs.

Typhoon attacks on the Korean Peninsula have recently become more frequent, and the strength of these typhoons is also gradually increasing because of climate change. Typhoon attacks cause storm surges in coastal regions; therefore, forecasts that enable advanced preparation for these storm surges are important. Because storm surge forecasts require both accuracy and speed, this study uses an artificial neural network algorithm suitable for nonlinear modeling and rapid computation. A storm surge forecast model was created for five tidal stations on the Korea Strait (southern coast of the Korean Peninsula), and the accuracy of its forecasts was verified. The model consisted of a deep neural network and convolutional neural network that represent the two-dimensional spatial characteristics. Data from the Global Forecast System numerical weather model were used as input to represent the spatial characteristics. The verification of the forecast accuracy revealed an absolute relative error of ≤5% for the five tidal stations. Therefore, it appears that the proposed method can be used for forecasts for other locations in the Korea Strait. Furthermore, because accurate forecasts can be computed quickly, the method is expected to provide rapid information for use in the field to support advance preparation for storm surges.