• Journal of Korea Water Resources Association
• /
• v.56 no.8
• /
• pp.471-484
• /
• 2023

Deep learning models based on generative adversarial neural networks are specialized in generating new information based on learned information. The deep generative models (DGMR) model developed by Google DeepMind is an generative adversarial neural network model that generates predictive radar images by learning complex patterns and relationships in large-scale radar image data. In this study, the DGMR model was trained using radar rainfall observation data from the Ministry of Environment, and rainfall prediction was performed using an generative adversarial neural network for a heavy rainfall case in August 2021, and the accuracy was compared with existing prediction techniques. The DGMR generally resembled the observed rainfall in terms of rainfall distribution in the first 60 minutes, but tended to predict a continuous development of rainfall in cases where strong rainfall occurred over the entire area. Statistical evaluation also showed that the DGMR method is an effective rainfall prediction method compared to other methods, with a critical success index of 0.57 to 0.79 and a mean absolute error of 0.57 to 1.36 mm in 1 hour advance prediction. However, the lack of diversity in the generated results sometimes reduces the prediction accuracy, so it is necessary to improve the diversity and to supplement it with rainfall data predicted by a physics-based numerical forecast model to improve the accuracy of the forecast for more than 2 hours in advance.

• Journal of Environmental Impact Assessment
• /
• v.25 no.6
• /
• pp.414-431
• /
• 2016

There are numerous closed and abandoned mines in Korea, from which diverse heavy metals (e.g., As, Cd, Cu, Pb, Zn) are released into the surrounding soil, groundwater, surface water, and crops, potentially resulting in detrimental effects on the health of nearby residents. Therefore, we performed human risk assessments of two abandoned metal mines, Yanggok (YG) and Samsanjeil (SJ). The exposure parameters used in this assessment were specific to residents near mines and the included exposure pathways were relevant to areas around metal mines. The computed total excess carcinogenic risks for both areas exceeded the acceptable carcinogenic risk ($1{\times}10^{-6}$), indicating that these areas are likely unsafe due to a carcinogenic hazard. In contrast, the non-carcinogenic risks of the two areas differed among the studied receptors. The hazard indices were higher than the unit risk (=1.0) for male and female adults in YG and male adults in SJ, suggesting that there are non-carcinogenic risks for these groups in the study areas. However, the hazard indices for children in YG and female adults and children in SJ were lower than the unit risk. Consumption of groundwater and crops grown in the area were identified as major exposure pathways for carcinogenic and non-carcinogenic hazards in both areas. Finally, the dominant metals contributing to carcinogenic and non-carcinogenic risks were As and As, Cu, and Pb, respectively. In addition, the carcinogenic and non-carcinogenic risks of YG were evaluated to be 10 and 4 times higher than those of SJ, respectively, resulted from the relatively higher exposure concentration of As in groundwater within SJ area. Because of lacking of several exposure parameters, some of average daily dose (ADD) could not be computed in this study. Furthermore, it is likely that the ADDs of crop-intake pathway included some errors because they were calculated using soil exposure concentrations and bioconcentration factor (BCF) rather than using crop exposure concentrations.

• Journal of Korea Port Economic Association
• /
• v.31 no.4
• /
• pp.75-89
• /
• 2015

This study examines 31 import and export cargo items handled in each port to investigate which items face the most competition among the ports and how many of them are transited to other ports. The study aims to suggest implications for the future port policy of Incheon Port. It was found that the volume concentration in the Western Coast region from 2005 to 2014 became increasingly decentralized. The decentralization began in earnest in 2009 in particular, and the value was 0.448 in 2014, indicating fierce competition among the regions. According to the static and dynamic positioning analyses results for Incheon Port, Pyeongtaek and Dangjin Port, and Gunsan Port, using BCG Matrix, the static positioning analysis showed that Incheon Port belongs to the 3rd quadrant (Cash Cows), Pyeongtaek and Dangjin Port belongs to the 2nd quadrant (Question Marks), and Gunsan Port belongs to the (Dogs) group. This implies that Incheon Port has maintained its position with large shares compared to those of other ports, despite its low growth rate. However, the market position and growth rate of Incheon Port decreased according to the dynamic positioning analysis results. The shift-share analysis results indicated that the volumes of Incheon Port and Gunsan Port were shifting to Pyeongtaek and Dangjin Port. Moreover, the ratio of absolute growth to potential growth of Incheon Port and Gunsan Port turned out to be significantly lower than that of Pyeongtaek and Dangjin Port, implying that Incheon Port and Gunsan Port are declining as compared to Pyeongtaek Port and Dangjin Port. According to the LQ index analysis results, specialized items from Incheon Port that do not overlap with other ports included the following ten items: meat, fish and crustaceans, bituminous coals, crude oil and petroleum, petroleum-refined products, plastic rubber and products, textiles, nonferrous metal and products, electric machinery, and aircrafts and ships. In particular, it was confirmed that the bulk cargo of Incheon Port was actually shifting to Pyeongtaek and Dangjin Port following the policy of re-establishing port functions.

• Journal of agricultural medicine and community health
• /
• v.35 no.3
• /
• pp.287-300
• /
• 2010

Objectives: We identified factors associated with any participation and with "good participation" (as assessed by frequency of attendance) in health promotion programs at a public health center in Korea. Methods: The subjects included 199 women who attended a baseline examination of health promotion programs during the first half of 2009. We collected data by structured interviews and physical examinations. Participation status was quantified by the frequency of attendance to the program. We classified the subjects as non-participants (0) and participants (1+ times), and as poor participants (0-29 times) and good participants (30+ times). Results: Of the 199 subjects, there were 57 (28.6%) non-participants, while 56 (28.1%) were classified as good participants. The factors that significantly affected participation status, as identified by univariate analysis, were personal factors (age, educational level, marital status, religion, living with someone, monthly income), environmental factors (method of access, accessibility of other facilities), body mass index, hypertension, perceived barriers to health, emotional salience, affectionate domain of social support, and depression. Multiple logistic regression analyses indicated that method of access was the most significant factor affecting participation in the health promotion program, and that the factors most highly associated with good participation were emotional salience, hypertension and body mass index. Conclusions: Our findings suggest that specific factors determine and enhance participation in health promotion programs offered by public health centers. These factors should be considered during the design and evaluation of health promotion programs that are offered by public health centers.