• Journal of Korea Water Resources Association
• /
• v.45 no.11
• /
• pp.1131-1141
• /
• 2012

In this study, factors that influence water supply capacities of dams are extracted using factor analysis, and multiple regression equations for estimating water supply capacities of dams are developed using the analysis results. Twenty-one multi-purpose dams and twelve Municipal and Industrial (M&I) water supply dams are selected for case studies, and eight variables influencing water supply capacities of dams, namely: watershed area, inflow, effective reservoir storage, grade on amount of M&I water supply, grade on amount of agricultural water supply, grade on amount of in-stream flow supply, grade on river administration, and grade on average rainfall, are determined. Two case studies for multi-purpose dams and M&I water supply dams are performed, employing factor analysis, respectively. For the two cases, preliminary tests, such as reviewing matrix of correlation coefficient, Bartlett's test of sphericity, and Kaiser-Meyer-Olkin (KMO) test, are conducted to evaluate the suitability of the variables for factor analysis. In case of multi-purpose dams, variables are grouped into three factors; M&I water supply dams, two factors. The factors are rotated using Varimax method, and then factor loading of each variable is computed. The results show that the variables influencing water supply capacities of dams are reasonably selected and appropriately grouped into factors. In addition, multiple regression equations for predicting the amounts of annual water supply of dams are established using the factor scores as explanatory variables, it is identified that the models' accuracies are high, and their applications to determining effective storage capacity of a dam during dam planning and design steps are presented. Consequently, it is thought that the variables and factors are useful for dam planning and dam design.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.11
• /
• pp.324-334
• /
• 2019

This study investigates the characteristics of patients hospitalized at a convalescent hospital, by considering patient groups based on their medical record information. It further analyzes and correlates the factors, thus providing basic data required to improve the patient classification system at convalescent hospitals. The data includes total information of 213 medical records of patients discharged from a convalescent hospital in the Jeonbuk region during a period of one year, from January to December 2016. The study examines and correlates the days of hospitalization and the number of diseases, revealing a positive correlation having a correlation coefficient, thereby indicating that infliction with a greater number of diseases results in longer hospitalization. Based on these findings, the study raises the need to revise and supplement items on the patient assessment report to help determine the patient groups and identify medical efforts to be actually provided to patients. In addition, a proper care service system for each patient group based on their respective problems that are intensively managed according to the patient groups will be an essential element in the efficient management of convalescent hospitals. Furthermore, an important task addressed will be in managing the health of the elderly population at the national level.

• Atmosphere
• /
• v.15 no.2
• /
• pp.75-90
• /
• 2005

This study examines the impacts of land cover changes on the East Asia summer monsoon with the National Center for Atmospheric Research Regional Climate Model (NCAR RegCM2), coupled with Biosphere Atmosphere Transfer Scheme (BATS). To assess the goals, two types of land cover maps were used in the simulation of summer climate. One type was NCAR land cover map (CTL) and the other was current land cover map derived from satellite data (land cover: LCV). Warm and cold surface temperature biases of $1-3^{\circ}C$ occurred over central China and Mongolia in CTL. The model produced excessive precipitation over northern land area but less over southern ocean of the model domain. Changes of biophysical parameters, such as albedo, minimum stomatal resistance and roughness length, due to the land cover changes resulted in the alteration of land-atmosphere interactions. Latent heat flux and wind speed in LCV increased noticeably over central China where deciduous broad leaf trees have been replaced by mixed farm and irrigated crop. As a result, the systematic warm biases over central China were greatly reduced in LCV. Strong cooling of central China decreased pressure gradient between East Asian continent and Pacific Ocean. The decreased pressure gradient suppressed the northward transport of moisture from south China and South China Sea. These changes reduced not only the excessive precipitation over north China and Mongolia but also less precipitation over south China. However, the land cover changes increased the precipitation over the Korean Peninsula and the Japan Islands, especially in July and August.

• Economic and Environmental Geology
• /
• v.42 no.6
• /
• pp.591-598
• /
• 2009

We made 2511 thermal property measurements on igneous, metamorphic, and sedimentary rock samples from Korea. The average thermal conductivities of igneous, metamorphic, and sedimentary rocks are 3.10 W/m-K, 3.76 W/m-K, and 3.54 W/m-K, respectively. Igneous rock can be classified into pluton, hypabyssal rock, and volconic rock; the average thermal conductivities of those rock types are 3.16 W/m-K, 3.26 W/m-K, and 2.77 W/m-K, respectively. Nonclastic sedimentary rock has higher thermal conductivity than clastic sedimentary rock. Thermal conductivity of Palezoic era rock is higher than Mesozoic era rock, because dominant mineral contents play an important role in the determination of thermal conductivity. Thermal conductivity of rocks is influenced by porosity. Therefore thermal conductivity of sedimentary rocks generally decreases with increasing porosity. Thermal conductivity and thermal diffusivity show linear correlation, its correlation coefficient of igneous, metamorphic, and sedimentary rocks are 0.775, 0.855, and 0.876, respectively.

• The Journal of Engineering Geology
• /
• v.30 no.3
• /
• pp.327-345
• /
• 2020

The effects of ground vibration on lava tubes during construction were studied to aid design of management and preservation measures for lava tubes. Ground conditions were assessed by RMR (Rock mass rating) and Q-system classifications for the Geomunoreum lava tubes, and vibration velocity was measured during in situ blasting tests in the Manjanggul and Yongcheondonggul lava tubes. Results indicate that the higher the rock quality, the greater the effect of vibration, although there is no clear linear relationship due to ground heterogeneity. A relationship derived between vibration velocity (PPV) and intensity (dB(V)) on the basis of blasting tests indicates that a vibration level of < 0.285 cm/sec meets the regulatory limit of 0.371 cm/sec and 65 dB(V) during daytime, and 0.285 cm/sec and 60 dB(V) during night. For blasting vibrations, square- and cube-root scaled distances are linearly correlated, with R² ≥ 0.76. On the basis of this correlation, explosive-charge weights meeting the 0.2 cm/sec vibration criterion for cultural heritage were estimated to be 2.88 kg at 50 m distance, and 11.52 kg at 100 m.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.39 no.6
• /
• pp.67-75
• /
• 2011

This study was carried out to investigate the changes in citizen satisfaction with parks and green spaces in Daegu. Three surveys were conducted using the same measurement tools in 1986, 2001, and 2010. Major results are as follows: 1. The overall satisfaction level with parks and green spaces increased remarkably over the last25years. Satisfaction levels with city spaces overall were higher than that of nearby surrounding areas. 2. Diversity of green spaces was the most distinguished indicator in increasing level of satisfaction while the management level of green space facilities showed the lowest improvement. 3. The factor analysis for individual variables for satisfaction resulted in two factors: functions and physical conditions were categorized as one factor, and indicators for the planning of green spaces were the other. Using a regression model, the major variables found for satisfaction were diversity, management level of woods, quantitative level, function of static recreation, and management level of facilities, respectively. 4. Regarding satisfaction level by the type of green spaces, green spaces by streets showed a remarkable increase while green spaces in industrial areas showed the lowest improvement. A factor analysis for each type of green space resulted in 3 factors: green spaces of urban parks, green spaces of urban recreational facilities, and green spaces of each district including residential areas and industrial areas.

• Korean Journal of Ecology and Environment
• /
• v.36 no.1 s.102
• /
• pp.29-37
• /
• 2003

Benthic macro-invertebrates were seasonally collected in the Onchen Stream in Pusan, from July 2001 to March 2002. Generally 4 phylum 5 class 10 order 19 family 23 species were observed in the study sites. Ephemeroptera, Plecoptera and various species appeared in headwater stream while Oligochaeta and Chironomidae were dominated in downstream sites. Community abundance patterns, especially the dominant taxa, Oligochaeta and Chironomidae, appeared to be different depending upon the sampling months. Oligochaeta was usually observed in July, December and March while Chironomidae was appeared in September. The biological indices, TBI(Trent Biotic Index), BS (Biotic Score), BMWP (Biological Monitoring Working Party)were calculated with the appeared communities of the sampling sites through the survey months. TBI showed 1 to 8, BMWP was 1 to 93 and CBI appeared 9 to 387 in the different sites. The biological indices decreased from headstream to downstream sites, We implemented the unsupervised Kohonen network for patterning of community abundance of the sampling sites. The patterning map by the Kohonen network was well represented community abundance of the sampling sites. Also, we conducted RTRN (Real Time Recurrent Neural Network) for predicting of the biological indices in the different sites. The results appeared that the predicting values by RTRN were well matched field data (correlation coefficient of TBI, BMWP and CBI were 0.957, 0.979 and 0.967, respectively).

• Journal of Forest and Environmental Science
• /
• v.15 no.1
• /
• pp.1-9
• /
• 1999

Noticing the intrinsic growing habit variations of hardwood species which is indigenous to the natural deciduous forest, eighteen tree species in overstory were selected for comparative evaluating tree form so as to develop the tree form index (TFI). Selected six tree form attributes were arbitrarily divided four assessment criteria and given appropriate scores. Eighteen tree species were undergone corresponding scores, converted to percentage base, and estimated TFI by the summation of the six scoring values. Cluster analysis was carried out to review which attribute would have been shared among species based on dissimilarity of scores for each pair of species in $18{\times}6$ data matrix. The result showed that Populus davidiana had the highest TFI value of 80.8 and Carpinus cordata had the lowest TFI va1ue of 46.3. The species which received more than 70 of TFI are Populus davidiana, Betula costata, Fraxinus mandshurica, and Ulmus davidiana var. japonica, characterized by the advantage of straight and longer bole, the narrower crown, and the thinner branch. On the contrary, such species as Sorbus alnifolia. Prunus sargentii, Acer mandshuricum, Juglans mandshurica, and Carpinus cordata received less than 60 of TFI, characterized by the disadvantage of the crooked or forked stem. the wider or deeper crown. and the thicker branch. In the dendrogram produced by cluster analysis, arbitrary value of Euclidean distance 10 divided eighteen species into four distinctive groups. and the typical characteristics of each group were discussed.

• Childhood Kidney Diseases
• /
• v.12 no.2
• /
• pp.150-156
• /
• 2008

Purpose : Many results have reported a correlation between the spot urine protein/creatinine ratio(P/C ratio) and 24-hour urinary protein(24UP) amount. This study was designed to evaluated correlation between 24UP amounts and P/C ratio in children and to find the factors that affect this correlation. Methods : 210 patients who visited the Department of Pediatrics in Busan Paik Hospital from september 2003 to december 2007 were included in this study. All the patients were divided into I, II, III/A, B, C group[I:24UP(mg/$m^2$/day)]<100, II: 100$\leq$24UP<1,000, III: 24UP$\geq$1,000, A: Cr excretion(mg/kg)<15, B: 15$\leq$Cr excretion<25, C: Cr excretion$\geq$25)]. Pearson correlation analysis was performed between 24UP and P/C ratio to evaluate the relationship. We defined fractional difference between 24UP and P/C ratio, and then performed multiple regression analysis. Results : There was a strong positive linear correlation between 24UP and P/C ratio in all patients, and the correlation was also good in each group. The factors affecting accurate quantitation of proteinuria using spot urine P/C ratio was creatinine excretion. Conclusion : Spot urine P/C ratio is a useful test to predict proteinuria roughly. Therefore, we expect that urine P/C ratio can be used as parameter instead of 24UP, if we set cutoff value of P/C ratio considered to creatinine excretion according to age and sex in large pediatric population.

• Korean Journal of Remote Sensing
• /
• v.32 no.5
• /
• pp.471-482
• /
• 2016

This paper presents an extraction method for water body which uses block-based image partitioning and extension of water body boundaries to improve the performance of supervised classification for water body extraction. The Mahalanobis distance image is created by computing the spectral information of Normalized Difference Water Index (NDWI) and Near Infrared (NIR) band images over a training site within the water body in order to extract an initial water body area. To reduce the effect of noise contained in the Mahalanobis distance image, we apply mean curvature diffusion to the image, which controls diffusion coefficients based on connectivity strength between adjacent pixels and then extract the initial water body area. After partitioning the extracted water body image into the non-overlapping blocks of same size, we update the water body area using the information of water body belonging to water body boundaries. The update is performed repeatedly under the condition that the statistical distance between water body area belonging to water body boundaries and the training site is not greater than a threshold value. The accuracy assessment of the proposed algorithm was tested using KOMPSAT-2 images for the various block sizes between $11{\times}11$ and $19{\times}19$. The overall accuracy and Kappa coefficient of the algorithm varied from 99.47% to 99.53% and from 95.07% to 95.80%, respectively.