• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.707-723
• /
• 2022

Although satellite-based sea surface temperature (SST) is advantageous for monitoring large areas, spatiotemporal data gaps frequently occur due to various environmental or mechanical causes. Thus, it is crucial to fill in the gaps to maximize its usability. In this study, daily SST composite fields with a resolution of 4 km were produced through a two-step machine learning approach using polar-orbiting and geostationary satellite SST data. The first step was SST reconstruction based on Data Interpolate Convolutional AutoEncoder (DINCAE) using multi-satellite-derived SST data. The second step improved the reconstructed SST targeting in situ measurements based on light gradient boosting machine (LGBM) to finally produce daily SST composite fields. The DINCAE model was validated using random masks for 50 days, whereas the LGBM model was evaluated using leave-one-year-out cross-validation (LOYOCV). The SST reconstruction accuracy was high, resulting in R² of 0.98, and a root-mean-square-error (RMSE) of 0.97℃. The accuracy increase by the second step was also high when compared to in situ measurements, resulting in an RMSE decrease of 0.21-0.29℃ and an MAE decrease of 0.17-0.24℃. The SST composite fields generated using all in situ data in this study were comparable with the existing data assimilated SST composite fields. In addition, the LGBM model in the second step greatly reduced the overfitting, which was reported as a limitation in the previous study that used random forest. The spatial distribution of the corrected SST was similar to those of existing high resolution SST composite fields, revealing that spatial details of oceanic phenomena such as fronts, eddies and SST gradients were well simulated. This research demonstrated the potential to produce high resolution seamless SST composite fields using multi-satellite data and artificial intelligence.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.5 no.3
• /
• pp.224-232
• /
• 2000

Inorganic nutrient concentrations in relation to springtime physical parameters of the Yellow Sea were investigated during April 1996. Three major water masses, i.e., the Yellow Sea Warm Current Water (YSWC), Coastal Current Water (CCW) and Changjiang River Diluted Water (CRDW), prevailed in the study area. Water masses were vertically wel1 mixed throughout the study area, and nutrients were supplied adequately from bottom to surface layer. As result of ample nutrients supplied by vertical mixing together with progressed daylight condition, springtime phytoplankton blooms were observed, which was responsible for the depletion of inorganic nutrients in surface water column. Low nutrients concentration in bottom water of the central Yellow Sea (Stn. D9; nitrate: <2 ${\mu}$M, phosphate: <0.3 ${\mu}$) was associated with the entrance of YSWC which is characterized by high temperature and salinity. Influenced by runoff and vertical tidal mixing, CCW with high nutrient concentrations probably associated with China and Korea coastal waters with high nutrients concentration. For the local scale of inorganic nutrient distribution, nutrient transfers from coast to central areas were limited due to restriction imposed by tidal fronts (Stn. D6) and thus affected the horizontal nutrient profiles. Relatively high phytoplankton biomass was observed in the tidal front (Chl-${\alpha}$=12.38 ${\mu}$gL$^{-1}$) during the study period. Overall, the springtime nutrient distribution patterns in the Yellow Sea appeared to be affected by: (1) Large-scale influx of YSWC with low nutrient concentrations and CCW with high nutrient concentrations influenced by Korea and China coastal waters; (2) vertical mixing of water mass and phytoplankton distribution; and (3) local-scale tidal front as well as phytoplankton blooms alongthe tidal front.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.7 no.2
• /
• pp.68-77
• /
• 2002

We investigated the outbreak, maintenance, and decline of the red tide dominated by C. polykrikoides in the coastal waters off Southern Korea from August to October, 2000, by combining field data and NOAA satellite images. In general, the C. polykrikoides blooms, which have occured annually in Korean coastal waters from 1995 to 1999, initiate between late August and early September around Narodo Island and expand to the whole area of the southern coast. However, initiation and short-term change of the bloom of 2000 were quite different from the pattern observed previously. In mid-August, thermal fronts in sea surface temperature(SST) were formed: 1) between the Tsushima Warm Current Water (TWCW) and the Southern Korean Coastal Waters (SKCW), 2) between the jindo cold water mass and the southwestern coastal waters, and 3) between the upwelled cold waters in the southeast coast and the offshore warm waters. Free-living cells of C. polykrikoides were concentrated in these frontal regions. In late August, the thermal front TWCW-SKCW approached the mouth of Yeosuhae Bay where Seomjin River water and anthropogenic pollutants from the Industrial Complex of Gwangyang Bay are discharged. In the blooms of 2000 initiated in Yeosuhae Bay in late August, the dominant species, C. polykrikoides, co-occured with Alexandrum tamarense, Gymnodinium mikimotoi, Skeletonema coastatum, and Chaetoceros spp. Two typhoons, 'Prapiroon' and 'Saomai' during and the C. polykrikoides bloom probably affected the abundance of this species. After the former typhoon passed the Korean Peninsula, cell growth of C. polykrikoides was maximal, but after the latter typhoon, the C. polykrikoides bloom disappeared (20 September). On 5 October, the blooms dominated by C. polykrikoides broke out within the coastal waters of Jinhae Bay and Hansan-Keoje Bay. NOAA satellite images showed that the isothermal line of 22$^{\circ}C$ extended into Jinhae Bay. In this bloom, C. polykrikoides also occurred simultaneously with Akashiwo sanguinea(=Gym-nodinium sangunium), a common red tide-forming dinoflagellate species in fall and winter in these coastal bays.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.14 no.4
• /
• pp.239-252
• /
• 1981

The East China Sea is an important region as nursery and spawning grounds for pelagic fishes such as jack mackerel, common mackerel etc. , and thus constitutes a major fishing area for purse-seine fishery. The environment surrounding in this region is under the influence of the Yellow Sea Cold Water, China Coastal Water and Kuroshio Current. The purpose of this study was to clarify the effects of oceanographic conditions and thermal fronts on the formation of the fishing grounds for the mackerels in the East China Sea. Through the analyses of fisheries statistics during 1968-1976 and temperature data, the following facts are found: 1) Approximately $70\%$ of the total mackerel(common) catches appeared to be come from the Tsushima Current region which includes Sakai coast of the Japan Sea, eastern Tsushima and Shirase Island, and Jeju Island of Korea. This area covers only about $8\%$ of the East China Sea. 2) Main fishing grounds for the jack mackerel are also centered around the area of southwestern Goto, Shirase and eastern Tsushima Island where the catches accounted for about $54\%$ of the total jack mackerel catches. 3) Fluctuations in annual catches are relatively small in the Tsushima Current region, compared to other regions such as Yellow Sea, southwestern coast of Kyushu and mid-western part of the East China Sea, where the fisheries yields varied considerably due to unstable fishing conditions. 4) It appears that the fishing grounds for the jack mackerel are mainly distributed along the warmer region ($15-20^{\circ}C$) of the thermal front, and those for the common mackerel are in somewhat colder region ($13-16^{\circ}C$) in the Tsushima Current.