한강하구 부유사의 공간적 분포

김영택(Young-Taeg Kim),김주연(Joo Youn Kim),김현성(Hyeon-Seong Kim),김국진(Kuk Jin Kim),박진영(Jin Young Park) 한국항해항만학회 2008 한국항해항만학회 학술대회논문집 Vol.2008 No.공동학술

Spatially averaged Suspended Sediment Concentration (SSSC) between the Shingok submerged dike and Jeonruri is 197 mg/l. In the vicinity of Ganghwado, Seokmodo, and Kyodongdo after Yudo, SSSC increases up to 374 mg/l, and decreases downstream to 151 mg/l at Hwangsando and to 25 mg/l at Palmido. Maximum SSSC occurs around Ganghwado and decreases seaward. In summary, it is believed that intermediate SSSC value occurs in the upper reach of Han River Estuary. The highest SSSC occurs in the middle, and lowest SSSC in the lower reach of the Han River Estuary.

한강하구 부유사의 공간적 분포

김영택(Young-Taeg Kim),김주연(Joo Youn Kim),김현성(Hyeon-Seong Kim),김국진(Kuk Jin Kim),박진영(Jin Young Park) 한국마린엔지니어링학회 2008 한국마린엔지니어링학회 학술대회 논문집 Vol.2008 No.-

Spatially averaged Suspended Sediment Concentration (SSSC) between the Shingok submerged dike and Jeonruri is 197 ㎎/ℓ. In the vicinity of Ganghwado, Seokmodo, and Kyodongdo after Yudo, SSSC increases up to 374 ㎎/ℓ, and decreases downstream to 151 ㎎/ℓ at Hwangsando and to 25 ㎎/ℓ at Palmido. Maximum SSSC occurs around Ganghwado and decreases seaward. In summary, it is believed that intermediate SSSC value occurs in the upper reach of Han River Estuary. The highest SSSC occurs in the middle, and lowest SSSC in the lower reach of the Han River Estuary.

다중빔 응향측심의 조석보정기법

김영택(Young-Taeg Kim),김주연(Joo-youn Kim),안영길(Young-Gil Ahn),한정식(Jung-Sik Han),김호균(Ho Kyun Kim),이은일(Eun-Il Lee) 한국마린엔지니어링학회 2008 한국마린엔지니어링학회 학술대회 논문집 Vol.2008 No.-

Corrections for multibeam soundings were reviewed to improve the method for tide correction. There are four correction methods for navigation, sounding, error, and tide. Among them, tide correction method was mainly investigated. Currently, four tide correction methods, such as constants for correction (time difference and height ratio for tide) and numerical tidal model are reviewed to identify the strengths and weakness of each method, and corresponding improving ways. They need to be carefully examined to identify advantages and disadvantages for the hybrid method to correct tide.

GAO model 적용성 검토

김영택(Young-Taeg Kim),이은일(Eunil Lee),백혜연(Hye yeon Baek),한충근(Chung-Keun Han),김국진(Kukjin Kim) 한국마린엔지니어링학회 2009 한국마린엔지니어링학회 학술대회 논문집 Vol.2009 No.-

Various methods for sediment transport paths are their applications reviewed to identify the potentials and limitations of each analysis tool. Gao and Collins (1992)'s two-dimensional trend method was applied to define the net sediment transport patterns in the southern channel of Jungangcheontae, Asan Bay. A revised code developed by Chang et al(2001), available at http://www.iamg.org/CGEditor/index.htm. was used as well.

멀티빔 음압자료를 이용한 저질분류

김영택(Young-Taeg Kim),이은일(Eunil Lee),박요섭(Yosup Park),권광석(kwang-seok kwon) 한국마린엔지니어링학회 2009 한국마린엔지니어링학회 학술대회 논문집 Vol.2009 No.-

To develop an automatic software which can remotely classify the seabed sediment type using multibeam echosounder backscatter data, a supervised classification method is implemented, which is composed of both PCA (Principal Component Analysis) for classifying the extracted pixels from the backscatter data and RGA (Regional Growing Algorithm) for clustering the selected pixels. Although basic technologies, such as acoustic and ground truth data acquisition, data processing and classification strategy, were independently achieved, the classification accuracy needs to be increased using geoacoustic studies and sediment texture-based classification methods.

다중빔 음향측심의 조선보정기법

김영택(Young-Taeg Kim),김주연(Joo-youn Kim),안영길(Young-Gil Ahn),한정식(Jung-Sik Han),김호균(Ho Kyun Kim),이은일(Eun-Il Lee) 한국항해항만학회 2008 한국항해항만학회 학술대회논문집 Vol.2008 No.공동학술

Corrections for multibeam soundings were reviewed to improve the method for tide correction. There are four correction methods for navigation, sounding, error, and tide. Among them, tide correction method was mainly investigated. Currently, four tide correction methods, such as constants for correction (time difference and height ratio for tide) and numerical tidal model are reviewed to identify the strengths and weakness of each method, and corresponding improving ways. They need to be carefully examined to identify advantages and disadvantages for the hybrid method to correct tide.

황해연안의 2013년 11월 이상조위편차 발생 원인

김호균,김영택,이동환,Kim, Ho-Kyun,Kim, Young Taeg,Lee, Dong Hwan 해양환경안전학회 2016 海洋環境安全學會誌 Vol.22 No.4

황해연안 조위관측소 10 개 지점에서 2013년 11월 24일 밤부터 25일 오전까지 관측한 해수면, 해면기압, 바람, 유동 자료뿐만 아니라 일기도를 분석하여 이상조위편차의 발생 원인과 관측자료들 간의 상호상관성을 알아보았다. 이상조위편차란 최대조위편차와 최소조위편차가 나타나는 시간동안 두 편차간의 차를 의미한다. 영종도의 최대조위편차는 111 cm, 최소조위편차는 -65 cm로, 4시간 1분 동안 176 cm의 이상조위편차를 보여 10개 조위관측소 가운데 가장 크다. 반면 모슬포의 이상조위편차는 8시간 52분 동안 약 68 cm로 가장 작다. 이 같은 이상조위편차는 기압점프에 의한 기상해일이 아니라 저기압에 의한 기압배치의 영향으로 바람에 의해 발생한 것으로 확인되었다. 각 지점에서 이상조위편차에 의한 흐름은 연평균 낙조류 세기의 16 ~ 41 %로 무시할 수 없을 정도이다. 조위편차, 바람, 조류잔차의 상호상관관계로부터 저기압의 중심이 한반도 서쪽에 가까이 위치해 있을 때 인천에서 남풍계열의 바람에 의한 북향류가 해수면을 상승시켰고, 한반도 통과 후 북풍계열의 바람에 의해 남향류가 해수면을 하강시켰다. The cause of abnormal tidal residuals was examined by analyzing sea levels, sea surface atmospheric pressures, winds at ten tide stations, and current, measured at the coast of the Yellow Sea from the night of November $24^{th}$ to the morning of the $25^{th}$ in 2013, along with weather chart. Additionally, the cross-correlations among the measured data were also examined. The 'abnormal tidal residuals' mentioned in this study refer to differences between maximum and minium tidal residuals. The largest abnormal tidal residual was identified to be a difference of 176 cm occurring over 4 hours and 1 minute at YeongJongDo (YJD) with a maximum tidal residual of 111 cm and minimum of -65 cm. The smallest abnormal tidal residual was 68 cm at MoSeulPo (MSP) during 8 hours 52 minutes. The cause of these abnormal tidal residuals was not a meteo-tsunami generated by an atmospheric pressure jump but wind generated by the pressure patterns. The flow speed due to these abnormal tidal residuals as measured at ten tide stations was not negligible, representing 16 ~ 41 % of the annual average ebb current speed. From the cross correlation among the tidal residuals, winds, and tidal residual currents, we learned the northern flow, due to southerly winds, raised the sea level at Incheon when a low pressure center located on the left side of the Korean Peninsula. After passing the Korean Peninsula, a southern flow due to northerly winds decreased the sea level.

비주석 재귀신경망 앙상블 모델을 기반으로 한 조위관측소 해수위의 준실시간 이상값 탐지

이은주,김영택,김송학,주호정,박재훈,LEE, EUN-JOO,KIM, YOUNG-TAEG,KIM, SONG-HAK,JU, HO-JEONG,PARK, JAE-HUN 한국해양학회 2021 바다 Vol.26 No.4

상시 관측되는 조위관측소 해수위 자료는 결측값과 오측값을 포함하고 있으며, 그 중 오측 값은 이상값으로 분류되는 전처리 대상이다. 이러한 오측을 제거하기 위해 대표적으로 3𝜎 (three standard deviations) 규칙이 적용되어왔으나, 기상이변 등에 의한 극값이 존재하거나 3𝜎 범위 안에서도 오측이 존재하는 해수위 자료에는 그 적용이 어렵다. 본 연구에서 설계된 모델은 오측에 대한 사전 정보가 필요하지 않은 비주석 학습으로 구성되며, 재귀신경망과 앙상블 기법을 이용함으로써 실시간으로 수집되는 해수위 자료가 오측일 가능성을 발생한지 20분 이내로 제시한다. 검증이 완료된 모델은 평시 및 기상이변시의 정상값과 오측값을 잘 분리하며, 학습이 이뤄지지 않은 연도의 해수위 자료에서도 이상값 탐지가 가능함을 확인하였다. 본 연구의 관측 이상치 탐지 알고리즘은 조위관측소 해수위에 국한되지 않고 다양한 해양 및 대기자료의 이상치 탐지 인공신경망 모델에 확장 적용할 수 있다. Real-time sea level observations from tide gauges include missing and erroneous values. Classification as abnormal values can be done for the latter by the quality control procedure. Although the 3𝜎 (three standard deviations) rule has been applied in general to eliminate them, it is difficult to apply it to the sea-level data where extreme values can exist due to weather events, etc., or where erroneous values can exist even within the 3𝜎 range. An artificial intelligence model set designed in this study consists of non-annotated recurrent neural networks and ensemble techniques that do not require pre-labeling of the abnormal values. The developed model can identify an erroneous value less than 20 minutes of tide gauge recording an abnormal sea level. The validated model well separates normal and abnormal values during normal times and weather events. It was also confirmed that abnormal values can be detected even in the period of years when the sea level data have not been used for training. The artificial neural network algorithm utilized in this study is not limited to the coastal sea level, and hence it can be extended to the detection model of erroneous values in various oceanic and atmospheric data.

한국연안 일평균 조위편차의 시공간적 변동 특성

김호균,김영택,Kim, Ho-Kyun,Kim, Young-Taeg 해양환경안전학회 2013 해양환경안전학회지 Vol.19 No.6

본 연구에서는 우리나라 연안의 2003~2009년 해수면자료로 조위편차를 산출하고, 일평균조위편차의 시공간적 변동을 EOF 분석, 해면기압과 바람이 조위편차 변동에 얼마나 영향을 미치는지를 상관성 분석을 통해 알아보았다. 일평균조위편차는 전체 변동량의 68 %(제1모드)가 동시승강하였고, 전체 변동량의 21 %(제2모드)는 서해안이 상승할 때 남해안과 동해안이 하강하는 교차승강을 하였다. 해역별로 조위편차에 영향을 주는 주요 요인을 보면, 서해안은 남-북 방향의 바람 성분이었고, 남해안은 동해안으로 갈수록 해면기압의 영향이 우세하였다. EOF analysis of tidal residual derived from 2003~2009 tide data was used to identify the spatio-temporal variability. The effect of sea surface air pressures and winds on the tidal residual was also investigated by the correlation analysis. The first mode accounting for 68 % of the total variance represented concurrent sea level rise or fall, and the second mode accounting for 21 % of the total variance explained alternative sea level rise and fall between West Sea coast and both South Sea and East Sea coasts. While northerly and southerly winds dominated the tidal residual in the eastern coast of Yellow Sea, the effect of sea surface air pressures on the tidal residual increased along the coastal regions from South Sea to East Sea.

한국 남부 해역 SST의 계절 및 경년 변동이 단기 딥러닝 모델의 SST 예측에 미치는 영향

주호정,채정엽,이은주,김영택,박재훈,JU, HO-JEONG,CHAE, JEONG-YEOB,LEE, EUN-JOO,KIM, YOUNG-TAEG,PARK, JAE-HUN 한국해양학회 2022 바다 Vol.27 No.2

Sea Surface Temperature (SST), one of the ocean features, has a significant impact on climate, marine ecosystem and human activities. Therefore, SST prediction has been always an important issue. Recently, deep learning has drawn much attentions, since it can predict SST by training past SST patterns. Compared to the numerical simulations, deep learning model is highly efficient, since it can estimate nonlinear relationships between input data. With the recent development of Graphics Processing Unit (GPU) in computer, large amounts of data can be calculated repeatedly and rapidly. In this study, Short-term SST will be predicted through Convolutional Neural Network (CNN)-based U-Net that can handle spatiotemporal data concurrently and overcome the drawbacks of previously existing deep learning-based models. The SST prediction performance depends on the seasonal and interannual SST variabilities around the southern coast of Korea. The predicted SST has a wide range of variance during spring and summer, while it has small range of variance during fall and winter. A wide range of variance also has a significant correlation with the change of the Pacific Decadal Oscillation (PDO) index. These results are found to be affected by the intensity of the seasonal and PDO-related interannual SST fronts and their intensity variations along the southern Korean seas. This study implies that the SST prediction performance using the developed deep learning model can be significantly varied by seasonal and interannual variabilities in SST.