ScienceWatch@KIOST
Not a member yet
29648 research outputs found
Sort by
Hybrid Reconstruction of Sea Level at Dokdo in the East Sea Using Machine Learning and Geospatial Interpolation (1993–2023)
No full textSea level variability in the East Sea (Sea of Japan) and the Northwest Pacific poses challenges for coastal risk management due to the scarcity of long-term observations at remote locations such as Dokdo (Dok Island). This study reconstructs a continuous monthly sea level record at Dokdo from 1993 to 2023 by imputing gaps in 13 nearby Permanent Service for Mean Sea Level tide gauge stations using eight machine learning models and geospatial interpolation methods. The ensemble mean of Machine Learning-based imputations produced physically realistic and temporally coherent timeseries, preserving both seasonal and interannual variability. Sea level at Dokdo, estimated via inverse distance weighting, aligned well with satellite altimetry from Copernicus Marine Service and exhibited strong regional coherence with nearby stations. These results demonstrate that a hybrid framework combining statistical imputation, Machine Learning, and inverse barometric correction can effectively reconstruct sea level in data-sparse marine regions. The methodology provides a scalable tool for monitoring long-term trends and validating satellite and model products in marginal seas like the East Sea.11Nsciescopu
Study on the detection of floating seaweeds(S. horneri) using GOCI-2B
No full text중국 동쪽 해안을 포함한 황해 및 동중국해는 세계 최대 규모의 해조류 대발생 해역 중 하나로 매년 녹조(green tide)와 갈조(gold tide)가 지속적으로 발생하고 있다. 연안에서 이탈한 해조류는 부유하면서 바람과 해류에 의해서 한반도 연안, 동해 및 일본 연안까지 유입되고 있다. 단일종으로 구성된 해조류 대발생은 20세기 초부터 보고되었으며 1970년대 이후에는 산업화 발달에 의한 연안 부영양화로 세계 곳곳에서 확대되고 있으며 그 규모도 점차 커지고 있다. 동중국해 및 황해 해역에서 발생하는 부유 괭생이모자반 대발생은 주로 모자반속(Sargassum)에 의해 발생하고 있다. 부유성 해조류 특히 외래기원 부유성 해조류인 괭생이모자반은 2008년부터 황해 및 동중국해에 존재해 왔지만, 최근 들어 발생량이 증가하면서 우리나라 연안에 많은 영향을 미치고 있다. 특히 2015년에 대규모의 모자반이 출현했고, 제주도와 신안 해안으로 대규모 유입이 발생했고, 최근 2020년과 2021년 대규모로 유입되면서 피해 증가되었다. 중국에서 유입된 괭생이모자반은 2015년 12,100톤, 2016년 19톤, 2017년 4,418톤, 2018년 2,150톤, 2019년 860톤, 2020년 5,186톤, 2021년 1월 5,913톤을 수거했다. 특히 괭생이모자반 수거에 소요된 예산은 2015년 800백만원, 2016년에는 해상에서 바로 수거가 이루어졌으며, 2017년 200백만원, 2018년에는 150백만원을 소요되었다. 본 연구에서는 GOCI 해색위성 자료를 이용하여 부유 해조류 탐지 결과을 기반으로 오탐지 원인에 대해서 분석했다. 그리고, 괭생이모자반 이동경향 분석을 위해 2008년부터 2025년까지 월별 탐지 결과를 기반으로 주제도 제작을 수행했다.2
Isolation and Characterization of Secondary Metabolites from the Marine Sponge Sarcotragus spinosulus
No full textThe marine sponge Sarcotragus spinosulus, collected from Chuuk, Micronesia, was investigated for its secondary metabolites. The study led to the isolation of irciniastatin A (1), a new-in-nature derivative (2), and a previously unreported derivative (3). In addition, eight meroterpenes were obtained: three hydroquinone derivatives (4–6), three chromene derivatives (7–9), and two chromane derivatives (10–11). An unreported 2-substituted thiazole bearing a saturated fatty acid side chain (12) was also characterized. Structural elucidation was performed using 1D- and 2D-NMR spectroscopy and confirmed by high-resolution mass spectrometry. Biological evaluation of all isolated compounds is currently in progress.1
Comparative study of acoustic characteristics and vocal patterns of small yellow croaker during two spawning seasons
No full textMany fish species produce sounds associated with spawning, feeding, and social interactions, and such acoustic signals reflect species-specific traits. Passive acoustic analysis has become a valuable approach for understanding fish behavior, supporting stock management,
and monitoring marine ecosystems. The small yellow croaker (Larimichthys polyactis), a key sound-producing species in the East China and Yellow Seas, was investigated during the 2024 spawning season in a coastal fish farm. That analysis revealed strong nocturnal calling activity
and the occurrence of synchronized vocalizations involving large numbers of individuals. Building on these findings, an additional recording was conducted during the 2025 spawning season at the same fish farm to evaluate year-to-year variability. Comparative analysis will
focus on acoustic characteristics and temporal patterns, providing insights into whether reproductive sound production remains consistent between years or shows variability influenced by ecological or environmental conditions. The results are expected to advance understanding of the spawning ecology of small yellow croaker and demonstrate the importance of continued passive acoustic monitoring as a tool for sustainable fisheries management and marine ecosystem assessment (This work was supported by the Korea Research Institute for
defense Technology planning and advancement (KRIT) grant funded by the Korean Government (DAPA; Defense Acquisition Program Administration) in 2022 (No. KRIT-CT- 22-056)).1
Hormonal and photoneuroendocrine regulation of ovarian maturation in the Japanese eel (Anguilla japonica)
No full textUnderstanding how endocrine and environmental signals coordinate reproductive maturation is essential for clarifying the complex physiology of the Japanese eel (Anguilla japonica), whose reproductive mechanisms remains only partly understood. This study investigated how estradiol, its receptors, and photoneuroendocrine modulators interact during experimentally induced ovarian maturation and related morphological changes. Plasma estradiol rose markedly through vitellogenesis, accompanied by increased expression of estrogen receptors in the eye, brain, pituitary, and ovary. Based on these findings, it is reasonable to suggest that estrogen may contribute to eye enlargement and enhanced retinal sensitivity, changes that likely support visual adaptation during oceanic spawning migration. Ovarian progression was further associated with stage-dependent transcription of gnrh1, fshβ, lhβ, and cyp19a, a decline in dopamine after late vitellogenesis, and a peak in melatonin secretion at final maturation. These results imply that melatonin conveys photic or lunar information to the reproductive axis, while dopamine modulates gonadotropin release, together refining the timing of oocyte development. Overall, the study highlights an integrated mechanism in which steroid hormones, retinal plasticity, and photoneuroendocrine pathways align reproductive physiology with environmental cues, providing insights useful for conservation and controlled maturation of A. japonica.1
Improvement of Wave Height Prediction Through Time-Series Decomposition and Convolutional LSTM
No full textThis study presents a deep learning-based approach to significant wave height prediction around the Korean Peninsula, integrating time-series decomposition techniques to enhance forecasting accuracy. While spatio-temporal deep learning models such as Convolutional LSTM (ConvLSTM) have been widely utilized in oceanic time-series predictions, they may be limited in effectively distinguishing complex temporal components. To address this, the Simple Ocean Prediction Network (SMOP-Net) was developed using a straightforward ConvLSTM-based architecture. Additionally, a new model, the Time-series Decomposition-based Ocean Prediction Network (TDOP-Net), was introduced to improve accuracy by incorporating decomposed time-series components—long-term trend, seasonal trend, and residuals—into the input data.
The models were trained and validated using ERA5 reanalysis data from 2016 to 2020, with performance evaluations conducted on 2021 data across the entire adjacent seas of Korea including the Yellow Sea, East China Sea, East Sea (Hersbach et al., 2020). Results indicate that TDOP-Net significantly outperforms SMOP-Net, particularly in nearshore areas where terrestrial influences are stronger. The application of time-series decomposition improved accuracy by up to 7% in Relative Root Mean Squared Error (RRMSE) and 29% in Relative Mean Absolute Error (RMAE), while also mitigating cumulative forecasting errors over extended lead times.
These findings demonstrate that integrating time-series decomposition into deep learning models enhances prediction stability and accuracy, offering a computationally efficient approach for operational wave forecasting. The proposed methodology can be extended to other oceanic and atmospheric predictions, contributing to improved coastal hazard preparedness and marine environmental management.1
Simulation of Gut Microbiota of Sebastes schlegelii Using Serially Connected Continuous Stirred Reactors
No full textAn appropriate simulation protocol has been increasingly necessary as it can reduce time and labor in studying intestinal microbiome and can be used to determine the effects of probiotic candidates. Simulator of the human intestinal microbial ecosystem (SHIME®) has been widely implemented to study the behavior of human intestinal microbiota. Nonetheless, such an approach has been hardly demonstrated in simulating other organisms besides human, especially ectothermic animals. In this study, we implemented the SHIME® approach for simulating intestinal microbiota of Sebastes schlegelii. The simulator consists of four bioreactors, which reflect stomach, pyloric caeca, anterior and posterior intestine. Microbial compositions in fecal and the samples taken from the intestinal reactors were found to be similar at the family level, indicating that the SHIME® approach can be implemented to simulate gut microbiota of other animals. However, our result suggests that further optimization will be necessary in that the portions of Vibrio and Photobacterium differed significantly between the fecal and the reactor samples. [The work was supported by Marine Biotics project (KIMST-20210649) and Ministry of Oceans and Fisheries (MOF), Republic of Korea]2
