• Journal of Korea Water Resources Association
• /
• v.56 no.4
• /
• pp.261-272
• /
• 2023

In recent years, natural disasters such as heavy rainfall and typhoons have occurred more frequently, and their severity has increased due to climate change. The Korea Meteorological Administration (KMA) currently uses the same criteria for all regions in Korea for watch and warning based on the maximum cumulative rainfall with durations of 3-hour and 12-hour to reduce damage. However, KMA's criteria do not consider the regional characteristics of damages caused by heavy rainfall and typhoon events. In this regard, it is necessary to develop new criteria considering regional characteristics of damage and cumulative rainfalls in durations, establishing four stages: blue, yellow, orange, and red. A classification model, called DSCM (Disaster Severity Classification Model), for the four-stage disaster severity was developed using four machine learning models (Decision Tree, Support Vector Machine, Random Forest, and XGBoost). This study applied DSCM to local governments of Seoul, Incheon, and Gyeonggi Province province. To develop DSCM, we used data on rainfall, cumulative rainfall, maximum rainfalls for durations of 3-hour and 12-hour, and antecedent rainfall as independent variables, and a 4-class damage scale for heavy rain damage and typhoon damage for each local government as dependent variables. As a result, the Decision Tree model had the highest accuracy with an F1-Score of 0.56. We believe that this developed DSCM can help identify disaster risk at each stage and contribute to reducing damage through efficient disaster management for local governments based on specific events.

• Tunnel and Underground Space
• /
• v.20 no.1
• /
• pp.15-27
• /
• 2010

Geo-hazard shows a rapid increasing tendency with establishment of frequent great slopes in various construction sites, especially in the unfavorable topographic condition in which about 70% of the surface is covered by the mountainous area. An repeatedly taking place on the heavy rain season is accompanied by a large scale of rockfall, and causes great damage to an individual as well as a property. Even though lots of field studies and fundamental studies have been performed to reduce this hazard, however, an essential study on the mechanism of the rockfall should be limited to the conventional studies on the slope reinforcement and/or the rockfall risk analysis. In this study, the mechanism of rockfall depending on the morphologic characteristics of slope has been simulated numerically with the PFC2D, one of the discrete element programs. For analyzing its mechanism, the input parameters relating to the slope such as surface condition, gradient, number of benches, bench gradient, and the ratio of bench width to rockfall size were taken into consideration.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.33 no.2 s.109
• /
• pp.71-82
• /
• 2005

A large scale of riverside rearrangement has been recently done in the major rivers in Korea. So inundation possibility in agricultural area closed by these rivers has been higher than the possibility a few years ago. However, land use in this area has not been adjusted to a change of this situation near the rivers. Therefore, when typhoon or heavy rain is happened on this area, it can cause a large damage in agricultural area. This study analyzed inundation potentiality in agricultural area at Ibang-myeon, Changnyeong-gun, Kyeongnam-province, Korea by using the logistic regression model and the piecewise regression model. The first thing we did was to transfer the inundation area per elevation to the accumulated inundation area per elevation. This accumulated inundation area per elevation as an distribution function could be described by the logistic regression model(LRM), and piecewise regression model(PRM) could make it much more accurate to analyze the inundation area per elevation. As a result, the regression models derived from LRM and PRM showed $R^2$ over 0.950. The models derived from LRM and PRM in Ibang-myeon noted that frequently inundated area(FIA) was shown up to 12.12m in elevation, and potentially inundated area(PIA) was shown up to 14.60m in elevation. In FIA, regular agricultural activity would be impossible. And It would be not easy to continue the regular agricultural activity in PIA. So, this land should be rearranged to be used for a buffer zone for ecosystem protection, landscape conservation and things like that in riverside.

• Journal of Korean Society for Geospatial Information Science
• /
• v.14 no.3 s.37
• /
• pp.87-95
• /
• 2006

Recently, local heavy rain for a short term is caused by unusual changing in the weather. This phenomenon has, several times, caused an extensive flash flood, casualties, and material damage. This study is aimed at calculating the characteristics of flash floods in streams. For this purpose, the analysis of topographical characteristics of water basin through applying GIS techniques will be conducted. The flash flood prediction model we used is made with GCIUH (geomorphoclimatic instantaneous unit hydrograph). The database is established by the use of GIS and by the extraction of streams and watersheds from DEM. The streams studied are included small, middle and large scale watersheds. For the first, for the establishment or criteria on the flash flood warning, peak discharge and trigger runoff must be decided. This study analyzed the degree or aptitude of topographical factors to the trigger runoff calculated by GCUH model.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.1
• /
• pp.153-158
• /
• 2016

A detection algorithm of fallen pear bags caused by natural disaster like heavy rain, typhoon, hurricane, etc. is presented in this paper. The algorithm is developed for the gray pear bags with printed characters which are widely used at pear farms at Sangju and Naju producing large quantity of pears for export. It sets a region of interest (ROI) at first and then eliminates the regions having chromatic color in ROI. Morphological operation and prior information are used to eliminate small noises and several unusual regions and finally the regions of fallen pear bags are remained. The remained regions are analyzed and counted to estimate the scale of damage. Test images are consisted of the images taken at pear farms of Sangju and Naju at 2014. Experimental result shows that the detection rate of pear bags is more than 90% and also the proposed system can be implemented in real-time using hand-held devices because of its simple and parallel architecture.

• Journal of Korea Water Resources Association
• /
• v.49 no.6
• /
• pp.519-528
• /
• 2016

Frequently torrential rain is occurred by climate change and urbanization. Urban is formed with road, residential and underground area. Without detailed topographic flooded analysis consideration can take a result which are wrong flooded depth and flooded area. Especially, flood analysis error of population and assets in dense downtown is causing a big problem for establishments and disaster response of flood measures. It can lead to casualties and property damage. Urban flood analysis is divided into sewer flow analysis and surface inundation analysis. Accuracy is very important point of these analysis. In this study, to confirm the effects of the elevation data precision in the process of flooded analysis were studied using 10m DEM, LiDAR data and 1:1,000 digital map. Study area is Dorim-stream basin in the Darim drainage basin, Sinrim 3 drainage basin, Sinrim 4 drainage basin. Flooding simulation through 2010's heavy rain by using XP-SWMM. Result, from 10m DEM, shows wrong flood depth which is more than 1m. In particular, some of the overflow manhole is not seen occurrence. Accordingly, detailed surface data is very important factor and it should be very careful when using the 10m DEM.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.408-417
• /
• 2016

The frequent regional torrential or heavy rain and typhoon mostly caused by climate change has resulted in sediment disasters particularly in mountainous or hilly areas. More than 65% of South Korea is mountainous and development and rapid urbanization has brought lots of steep sloping industrial complexes, which are adjacent to cities. Such continuous urbanization and industrialization can result in an increase in serious damage to those places. Korea has very high population density so sediment disaster could result in a tremendous loss of property and life. A recent 10-year (2001~2010) study of the average annual loss shows 68 casualties and property loss of 1.7044 trillion Won(?), which indicates a 20% and 25% decrease for both life and property, respectively, but urban areas are experiencing increasing damage. In this paper, a comprehensive simulator composed by references, analyses, and the recent technologies was applied to visualize the scale of the damaged Woomyeon-san (Mt.) and verify the performance of the simulator.