• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.603-614
• /
• 2021

Recently, the surface temperature in the seas around Korea has been continuously rising. This temperature rise causes changes in fishery resources and affects leisure activities such as fishing. In particular, high temperatures lead to the occurrence of red tides, causing severe damage to ocean industries such as aquaculture. Meanwhile, changes in sea temperature are closely related to military operation to detect submarines. This is because the degree of diffraction, refraction, or reflection of sound waves used to detect submarines varies depending on the ocean mixed layer. Currently, research on the prediction of changes in sea water temperature is being actively conducted. However, existing research is focused on predicting only the surface temperature of the ocean, so it is difficult to identify fishery resources according to depth and apply them to military operations such as submarine detection. Therefore, in this study, we predicted the temperature of the ocean mixed layer at a depth of 38m by using temperature data for each water depth in the upper mixed layer and meteorological data such as temperature, atmospheric pressure, and sunlight that are related to the surface temperature. The data used are meteorological data and sea temperature data by water depth observed from 2016 to 2020 at the IEODO Ocean Research Station. In order to increase the accuracy and efficiency of prediction, LSTM (Long Short-Term Memory), which is known to be suitable for time series data among deep learning techniques, was used. As a result of the experiment, in the daily prediction, the RMSE (Root Mean Square Error) of the model using temperature, atmospheric pressure, and sunlight data together was 0.473. On the other hand, the RMSE of the model using only the surface temperature was 0.631. These results confirm that the model using meteorological data together shows better performance in predicting the temperature of the upper ocean mixed layer.

• Journal of Korea Water Resources Association
• /
• v.54 no.spc1
• /
• pp.1023-1035
• /
• 2021

The electricity cost of a desalination facility was also predicted and reviewed, which allowed the proposed model to be incorporated into the future design of such facilities. Input data from 2003 to 2014 of the Korea Hydrographic and Oceanographic Agency (KHOA) were used, and the structure of the model was determined using the trial and error method to analyze as well as hyperparameters such as salinity and seawater temperature. The future seawater quality was estimated by optimizing the prediction model based on machine learning. Results indicated that the seawater temperature would be similar to the existing pattern, and salinity showed a gradual decrease in the maximum value from the past measurement data. Therefore, it was reviewed that the electricity cost for seawater desalination decreased by approximately 0.80% and a process configuration was determined to be necessary. This study aimed at establishing a machine-learning-based prediction model to predict future water quality changes, reviewed the impact on the scale of seawater desalination facilities, and suggested alternatives.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.29 no.2
• /
• pp.101-115
• /
• 2024

The ocean heatwave is emerging as a major issue due to global warming, posing a direct threat to marine ecosystems and humanity through decreased food resources and reduced carbon absorption capacity of the oceans. Consequently, the prediction of ocean heatwaves in the vicinity of the Korean Peninsula is becoming increasingly important for marine environmental monitoring and management. In this study, an LSTM model was developed to improve the underestimated prediction of ocean heatwaves caused by the coarse vertical grid system of the Korean Peninsula Ocean Prediction System. Based on the results of ocean heatwave predictions for the Korean Peninsula conducted in 2023, as well as those generated by the LSTM model, the performance of heatwave predictions in the East Sea, Yellow Sea, and South Sea areas surrounding the Korean Peninsula was evaluated. The LSTM model developed in this study significantly improved the prediction performance of sea surface temperatures during periods of temperature increase in all three regions. However, its effectiveness in improving prediction performance during periods of temperature decrease or before temperature rise initiation was limited. This demonstrates the potential of the LSTM model to address the underestimated prediction of ocean heatwaves caused by the coarse vertical grid system during periods of enhanced stratification. It is anticipated that the utility of data-driven artificial intelligence models will expand in the future to improve the prediction performance of dynamical models or even replace them.

• Ecology and Resilient Infrastructure
• /
• v.2 no.3
• /
• pp.198-205
• /
• 2015

Impacts of climate change on aquatic ecosystems range from changes in physiological processes of aquatic organisms to species distribution. In this study, MaxEnt that has high prediction power without nonoccurrence data was used to simulate fish distribution changes in the Geum river watershed according to climate change. The fish distribution in 2050 and 2100 was predicted with RCP 8.5 climate change scenario using fish occurrence data (a total of 47 species, including 17 endemic species) from 2007 to 2009 at 134 survey points and 9 environmental variables (monthly lowest, highest and average air temperature, monthly precipitation, monthly lowest, highest and average water temperature, altitude and slope). The fitness of MaxEnt modeling was successful with the area under the relative operating characteristic curve (AUC) of 0.798, and environmental variables that showed a high level of prediction were as follows: altitude, monthly average precipitation and monthly lowest water temperature. As climate change proceeds until 2100, the probability of occurrence for Odontobutis interrupta and Acheilognathus yamatsuatea (endemic species) decreases whereas the probability of occurrence for Microphysogobio yaluensis and Lepomis macrochirus (exotic species) increases. In particular, five fish species (Gnathopogon strigatus, Misgurnus mizolepis, Erythroculter erythropterus, A. yamatsuatea and A. koreensis) were expected to become extinct in the Geum river watershed in 2100. In addition, the species rich area was expected to move to the northern part of the Geum river watershed. These findings suggest that water temperature increase caused by climate change may disturb the aquatic ecosystem of Geum river watershed significantly.

• Korean Journal of Environment and Ecology
• /
• v.34 no.1
• /
• pp.55-62
• /
• 2020

To estimate headwater stream temperature with seasonal variations, we analyzed precipitation, runoff and air temperature in experimental forest of Kangwon National University, Gangwon-do (2017~2018 years). The daily mean value of headwater stream temperature for spring was 6.9~17.7℃ and correlated with air temperature, that for summer and fall were 12.2~26.3℃ and 3.6~19.3℃, correlated with air temperature and runoff. Based on seasonal variations, we applied for stepwise multiple linear regression analyses to estimate headwater stream temperature with seasonal variations. The equations were headwater stream temperature(WT)_spring=(0.553×Air temperature)+(0.086×Runoff)+4.145 (R²=0.505; p<0.01), WT_summer=(0.756×Air temperature)+(-0.072×Runoff)+2.670 (R²=0.510; p<0.01), and WT_fall=(0.738×Air temperature)+(0.028×Precipitation)+2.660 (R²=0.844; p<0.01). The coefficient of determination (R²) was greater than when it was estimated by air temperature in all seasons and progressively increased from spring to winter. Therefore, we indicated difference on estimated magnitude of stepwise multiple linear regression, due to effects on headwater stream temperature of different environmental factors with seasonal variations. Furthermore, temporal factors with spatial characteristics (e.g., river versus headwater stream) could be recommended for estimating headwater stream temperature.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.139-139
• /
• 2020

Intergovernmental Panel on Climate Change(IPCC)에 따르면 지난 1세기 반 동안 전 세계 평균 기온은 약 1℃가 상승하였으며, 온실가스 축적에 따라 평균기온은 21세기 중반에서 21세기 말까지 1~3℃가 증가할 것으로 전망되고 있다. 이러한 기온의 상승으로 인한 하천의 수온 변화는 수중에서 온도에 민감한 생화학적 반응의 변화를 유발하여 수질 및 수생태 변화에 영향을 미칠 수 있다. 따라서 효과적인 수질 및 수생태 관리를 위해서는 기온과 수온 사이의 명확한 관계 정립을 통해 수질변화를 정확하게 예측하는 것이 중요하다. 본 연구에서는 국내·외로 널리 활용되고 있는 SWAT(Soil and Water Assessment Tool, SWAT) 모형을 통해 기온-수온 회귀식이 하천 수질변화에 미치는 영향을 정량적으로 분석하고자 하였다. 그러나 기존 SWAT 모형에서의 기온-수온 회귀식은 미국 유역의 환경 특성을 바탕으로 도출되었기 때문에 국내 유역에 적용하기에 한계점이 있다. 따라서 본 연구의 목적은 국내 유역에서의 실측 기온자료와 수온자료를 사용하여 SWAT 모형 내 기온-수온 회귀식을 재도출하고 적용성을 평가하는 것이다.

• Journal of Korea Water Resources Association
• /
• v.46 no.1
• /
• pp.1-12
• /
• 2013

In this study, the potential effects of increased water temperature on fish habitat were analysed in the streams of Nakdong River watershed. The changes in suitable habitats for each fish species and in species number at a habitat site were predicted, based on the maximum thermal tolerances of 22 fish species. The estimated maximum thermal tolerance ranged between $27.7^{\circ}C$ and $33.1^{\circ}C$. Then, the increase of water temperature in 78-sites of Nakdong River watershed by 2100 was predicted by using the estimated air temperature data by 2100 in the literature and the regression analysis between air-temperature and water-temperature at each sites. The water temperature was estimated to have increased by $0.69^{\circ}C$, $1.76^{\circ}C$, and $2.32^{\circ}C$ in 2011~2040 (period S1), 2041~2070 (S2), and 2071~2100 (S3), respectively. With such increases in water temperature, the averaged suitable habitats for all 22 fish species would be influenced by 21.9%, 36.3%, and 51.4% in periods S1, S2, and S3, respectively.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.6
• /
• pp.238-245
• /
• 2021

In this study, we carried out case study to predict dissolved oxygen (DO) concentration of Nakdong river estuary with LSTM model. we aimed to figure out a optimal model condition and appropriate predictor for prediction in dissolved oxygen concentration with model parameter and predictor as cases. Model parameter case study results showed that Epoch = 300 and Sequence length = 1 showed higher accuracy than other conditions. In predictor case study, it was highest accuracy where DO and Temperature were used as a predictor, it was caused by high correlation between DO concentration and Temperature. From above results, we figured out an appropriate model condition and predictor for prediction in DO concentration of Nakdong river estuary.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.450-450
• /
• 2018

최근 기후변화와 수온상승으로 인한 녹조발생이 빈번하게 나타나며, 녹조발생에 관한 관심은 꾸준히 증가하고 있는 추세이다. 본 연구는 효율적인 녹조관리를 위하여 모델트리를 활용하여 클로로필-a 단기조류예측 기법을 개발하였다. 대상지역으로 영산강수계의 죽산보를 선정하였으며, 2013년 1월부터 2016년 12월까지 나주 수질자동측정망의 일 단위자료와 동일기간 광주 기상청의 일별 기상자료를 이용하였다. 상관 분석을 통해 T-N, T-P, N/Pratio와 클로로필-a, 수온, 일사량, 강수량을 독립변수로, 단기(t+1일, t+3일, t+5일, t+7일) 클로로필-a를 종속변수로 선정하여 단기조류예측기법을 개발하였다. 수집한 자료의 데이터세트는 격일 간격으로 Training, Testing 기간으로 구분하여 적용한 결과, 상관계수는 1일 예측 시, Training 기간에 0.89, Testing 기간에 0.91, 3일 예측 시, Training 기간에 0.74, Testing 기간에 0.68, 5일 예측 시, Training 기간에 0.70, Testing 기간에 0.66, 7일 예측 시, Training 기간에 0.63, Testing 기간에 0.62로 나타났다. RMSE(Root Mean Square Error)는 1일 예측 시, Training 기간에 13.96, Testing 기간에 12.22, 3일 예측 시, Training 기간에 20.03, Testing 기간에 22.14, 5일 예측 시, Training 기간에 21.32, Testing 기간에 22.57, 7일 예측 시, Training 기간에 23.52, Testing 기간에 23.45로 나타났다. 예측주기에 따라 모델트리와 회귀식에서 활용한 독립변수는 1일 예측 시, 모델트리는 N/Pratio, 클로로필-a, 회귀식은 클로로필-a로 다르게 나타났다. 반면, 3일, 5일, 7일 예측 시, 모델트리와 회귀식에 활용된 변수는 같게 나타났다. 클로로필-a, 수온, 일사량은 5일 예측 시 활용된 변수로, 3일 예측 시에는 기상항목인 강수량이, 7일 예측 시에는 수질항목인 T-N, N/Pratio가 추가되었다. 특히 1일 예측 시 일 때, 높은 예측정도와 활용된 변수의 수가 적게 나타나는 것을 확인하였으며, 예측기간이 길어질수록 예측의 정확성이 낮아지고, 활용된 변수의 수가 많아지는 것을 확인하였다. 향후 적정한 예측기간을 판단하고 예측가능성을 높이기 위해서는 지속적인 자료취득 및 개선이 필요하며, 이를 바탕으로 적절한 단기조류예측이 가능할 것으로 판단된다.

• Journal of the Korean earth science society
• /
• v.26 no.3
• /
• pp.268-275
• /
• 2005

One of the nonlinear statistical modelling, neural network method was applied to predict the Sea Surface Temperature Anomalies (SSTA) in the Nino regions, which represent El Nino indices. The data used as inputs in the training step of neural network model were the first seven empirical orthogonal functions in the tropical Pacific $(120^{\circ}\;E,\;20^{\circ}\;S-20^{\circ}\;N)$ obtained from the NCEP/NCAR reanalysis data. The period of 1951 to 1993 was adopted for the training of neural network model, and the period 1994 to 2003 for the forecasting validation. Forecasting results suggested that neural network models were resonable for SSTA forecasting until 9-month lead time. They also predicted greatly the development and decay of strong E1 Nino occurred in 1997-1998 years. Especially, Nino3 region appeared to be the best forecast region, while the forecast skills rapidly decreased since 9-month lead time. However, in the Nino1+2 region where they are relatively low by the influence of local effects, they did not decrease even after 9-month lead time.