• Journal of Korea Water Resources Association
• /
• v.50 no.7
• /
• pp.503-511
• /
• 2017

In order to estimate the demand for water resources planning and operation, methodology for determining the size of water supply facilities has been mainly applied to agricultural water, unlike living and industrial water, which reflects actual usage trends. This inevitably leads to an overestimation of agricultural water and can lead to an imbalance in the supply and demand of each use in terms of the total water resources plan. In this study, the difference of approaches of concept of net consumption was examined in comparison with the existing methodology and the characteristics of agricultural water demand were analyzed by applying it to whole Jeju Island. SWAT model was applied to estimate the amount of evapotranspiration, which is a key factor in estimating demand, and watershed modeling was performed to reflect geographical features, weather, runoff and water use characteristics of Jeju Island. For the past period (1992~2013), demand of Jeju Island as a whole was analyzed as 427 mm per year, and it showed a relatively high demand around the eastern and western coastal regions. Annual demand and seasonal variation characteristics of 10 river basins with watershed area of $30km^2$ or more were also analyzed. In addition, by applying the cultivated area of each crop in 2020 in the future, it is estimated that the demand corresponding to the 10-year frequency drought is 54% of the amount demanded in the previous research. This is due to the difference in approach depending on the purpose of the demand calculation. From the viewpoint of water resource management and operation, additional demand is expected as much as the net consumption. However, from the actual supply perspective, it can be judged that a facility plan that meets the existing demand amount is necessary. In order to utilize the methodologies and results presented in this study in practice, it is necessary to make a reasonable discussion in terms of policy and institutional as well as engineering verification.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.17 no.2
• /
• pp.151-156
• /
• 1984

The Chili fish, one of the most common species of the freshwater fishes, is distributed in the river basins of the western coast of Korea. However, any studies on this fish, especially its development or life history, have not been made so far. The authors, therefore, carried out an experiment to reveal some characteristics of egg develoment and fry in this fish. For accomplshing this study, we caught some adult Chili fish by means of a casting net at raceway in the vicinity of Kunsan in June and July, 1983; then, we injected 0. 1 cc Gonadotrophic hormone 1,000 unit solution (Puberogen) in order to induce the spawning on the abdominal cavity of male and female respectively, Approximately 12 hours after injection, the artificially fertilized eggs by using of the dry method and the naturally fertilized eggs were observed simultaneously in the aquarium ($30{\times}50{\times}70cm$). The fertilized eggs, in spherical shape, were sperative, demersel, adhesive nature and $1.38{\sim}1.59\;mm$ in diameter. The light yellow yolk was opaque and $1.17{\sim}l.27\;mm$ in diameter. The hatching of eggs took place at 29.5 hours after fertilization under the condition of $23.5{\sim}25.0^{\circ}C$ water temperature. Basically there were not found so many different aspects in comparison with the common species of Cyprinidae in the development process of eggs : however, there were a few noticable characteristics, that is, pigment cells, eye balls and blood circulation are invisible, or the velocity of egg development is rapid, seemed similar to the Loach's egg. Newly hatched larvae with the yolk sac (2.39 mm in size) were $3.87{\sim}3.99\;mm$ in total length and their myotome number was expressed as 28+15=43. About 20 hours after hatching, a caudal and a pectoral fin as well as an anus began to orgin. In addition, a few malanophore showed on the upper regin of the yolk sac of the larvae whose size was 4.73 mm in total length. The larvae developed about 5.02 mm in total length, and the median fin fold in the abdomen extended forward to two thirds or beyond of the yolk sac at 50 hours after hatching. Simultaneously, the eye balls and mouth also appeared. After 3 days, the yolk sac nearly disappeared, and the larvae (5.4 mm in total length) with perfect eye balls began to swim for themselves for the first time. Also, the specialization between the gill arches and the ailmentary canal, the development of each fins except dorsal fin as well as the blood circulation obviously distinguished in this stage. The larvae whose total length was 5.68 mm was very brisk: also, the development of the dorsal fin rudiment, nearly perfect formation of each organs and the function of the upper and lower jaws were noticeably observed 5 days after hatching.

• 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.