• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.12 no.3
• /
• pp.170-177
• /
• 2007

In this study, the long-term predictability of El Nino and La Nina events of Pusan National University Coupled General Circulation Model(PNU/CME CGCM) developed from a Research and Development Grant funded by Korea Meteorology Administration(KMA) was examined in terms of the correlation coefficients of the sea surface temperature between the model and observation and skill scores at the tropical Pacific. For the purpose, long-term global climate was hindcasted using PNU/CME CGCM for 12 months starting from April, July, October and January(APR RUN, JUL RUN, OCT RUN and JAN RUN, respectively) of each and every years between 1979 and 2004. Each 12-month hindcast consisted of 5 ensemble members. Relatively high correlation was maintained throughout the 12-month lead hindcasts at the equatorial Pacific for the four RUNs starting at different months. It is found that the predictability of our CGCM in forecasting equatorial SST anomalies is more pronounced within 6-month of lead time, in particular. For the assessment of model capability in predicting El Nino and La Nina, various skill scores such as Hit rates and False Alarm rate are calculated. According to the results, PNU/CME CGCM has a good predictability in forecasting warm and cold events, in spite of relatively poor capability in predicting normal state of equatorial Pacific. The predictability of our CGCM was also compared with those of other CGCMs participating DEMETER project. The comparative analysis also illustrated that our CGCM has reasonable long-term predictability comparable to the DEMETER participating CGCMs. As a conclusion, PNU/CME CGCM can predict El Nino and La Nina events at least 12 months ahead in terms of NIino 3.4 SST anomaly, showing much better predictability within 6-month of leading time.

• Journal of Wetlands Research
• /
• v.25 no.4
• /
• pp.306-314
• /
• 2023

Stream sediments are an important component of water quality management because they are receptors of various pollutants such as heavy metals and organic matters emitted from upland sources and can be secondary pollution sources, adversely affecting water environment. To effectively manage the stream sediments, identification of primary sources of sediment contamination and source-associated control strategies will be required. We evaluated the performance of machine learning models in identifying primary sources of sediment contamination based on the physico-chemical properties of stream sediments. A total of 356 stream sediment data sets of 18 quality parameters including 10 heavy metal species(Cd, Cu, Pb, Ni, As, Zn, Cr, Hg, Li, and Al), 3 soil parameters(clay, silt, and sand fractions), and 5 water quality parameters(water content, loss on ignition, total organic carbon, total nitrogen, and total phosphorous) were collected near abandoned metal mines and industrial complexes across the four major river basins in Korea. Two machine learning algorithms, linear discriminant analysis (LDA) and support vector machine (SVM) classifiers were used to classify the sediments into four cases of different combinations of the sampling period and locations (i.e., mine in dry season, mine in wet season, industrial complex in dry season, and　industrial complex in wet season). Both models showed good performance in the classification, with SVM outperformed LDA; the accuracy values of LDA and SVM were 79.5% and 88.1%, respectively. An SVM ensemble model was used for multi-label classification of the multiple contamination sources inlcuding landuses in the upland areas within 1 km radius from the sampling sites. The results showed that the multi-label classifier was comparable performance with sinlgle-label SVM in classifying mines and industrial complexes, but was less accurate in classifying dominant land uses (50~60%). The poor performance of the multi-label SVM is likely due to the overfitting caused by small data sets compared to the complexity of the model. A larger data set might increase the performance of the machine learning models in identifying contamination sources.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_1
• /
• pp.655-667
• /
• 2023

Algal bloom outbreaks are frequently reported around the world, and serious water pollution problems arise every year in Korea. It is necessary to protect the aquatic ecosystem through continuous management and rapid response. Many studies using satellite images are being conducted to estimate the concentration of chlorophyll-a (Chl-a), an indicator of algal bloom occurrence. However, machine learning models have recently been used because it is difficult to accurately calculate Chl-a due to the spectral characteristics and atmospheric correction errors that change depending on the water system. It is necessary to consider the factors affecting algal bloom as well as the satellite spectral index. Therefore, this study constructed a dataset by considering water quality, hydrological and meteorological factors, and sentinel-2 images in combination. Representative ensemble models random forest and extreme gradient boosting (XGBoost) were used to predict the concentration of Chl-a in eight weirs located on the Nakdong river over the past five years. R-squared score (R²), root mean square errors (RMSE), and mean absolute errors (MAE) were used as model evaluation indicators, and it was confirmed that R² of XGBoost was 0.80, RMSE was 6.612, and MAE was 4.457. Shapley additive expansion analysis showed that water quality factors, suspended solids, biochemical oxygen demand, dissolved oxygen, and the band ratio using red edge bands were of high importance in both models. Various input data were confirmed to help improve model performance, and it seems that it can be applied to domestic and international algal bloom detection.