• Journal of the Korean Geotechnical Society
• /
• v.30 no.12
• /
• pp.15-25
• /
• 2014

In this study, types of rock slope failure are analyzed by considering both joint characteristics investigated on previous landslide regions located at northern part of Mt. Jiri and geographic features of natural slopes deduced from GIS. The landslide prediction map was produced by superposing the frequency ratio layers for the six geographic features including elevation, slope aspect, slope angle, shaded relief, curvature and stream distance, and then the landslide risk map was deduced by combination of the prediction map and the damage map obtained by taking account of humanity factors such as roads and buildings in the study area. According to analysis on geographic features for previous landslide regions, the landslides occurred as following rate: 88% at 330~710 m in elevation, 77.7% at $90{\sim}270^{\circ}$ in slope aspect, 93.9% at $10{\sim}40^{\circ}$ in slope angle, 82.78% at grade3~7 in shaded relief, 86.28% at -5~+5 in curvature, and 82.92% within 400m in stream distance. Approximately 75% of the landslide regions belongs to the region of 'high' or 'very high' grade in the prediction map, and 13.27% of the study area is exposed to 'high risk' of landslide.

• Proceedings of the KSEEG Conference
• /
• 2007.04a
• /
• pp.322-324
• /
• 2007

• Proceedings of the KGS Conference
• /
• 2003.11a
• /
• pp.89-92
• /
• 2003

한반도에서는 장마전선이나 태풍으로 인하여 발생된 산사태가 큰 재해의 원인 중에 하나이다. 그러므로 산사태에 대하여 재해를 예측하거나 예방하기 위해서는 그 산사태 발생원인이 밝혀져야 한다. 그러나 산사태 발생원인을 수문학적으로 밝히려고 한 연구 사례는 국내에서는 거의 없다. 그래서 본 연구에서는 산사태를 발생시키는 주요한 영력인 산지유역사면에서 물의 움직임을 알아보기를 위해서 강우유출특성을 밝혔으며 그 특성의 차이에 미치는 토양구조의 영향을 나타냈다. (중략)

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.32 no.4_1
• /
• pp.281-292
• /
• 2014

This study examined the accuracy that produced using various types and combinations of landslide-related factors from landslide susceptibility index maps. A database of landslide-related factors was adopted by the landslide locations that obtained from aerial photographs, and the topographic factors that derived from airborne LiDAR observations and digital maps, and various soil, forest, and land cover. Landslide susceptibility index maps were calculated by logistic regression and frequency ratio from the landslide susceptibility index. The correlation between airborne LiDAR data and digital map was shown strong similarities with one another. Landslide susceptibility index maps indicated the existence of a strong correlation and high prediction accuracy, especially when the frequency ratio and airborne LiDAR were used. Therefore, we concluded that the Airborne LiDAR will contribute to the development of effective landslide prediction methods and damage reduction measures.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.12 no.2
• /
• pp.1-10
• /
• 2009

Recently the large scale of forest disaster such as landslide and forest fire gives a very bad impact on not only forest ecosystem but also farm business so that it has became the main issue of environmental problems. In this study, the landslide hazard area forecast method was developed by considering not only the topographic thematic maps based on GIS and satellite images but also amount of rainfall data, which are very important factors of landslide. Uljin-gun was selected as the study area and the GIS weight score and overlay analysis were applied to topographical map and meteorological observation map. Finally the landslide area distribution map was constructed by considering the evaluation criteria. Also, the accuracy could be acquired by comparing the landslide hazard area forecast map and real damaged area extracted from satellite image.

• Journal of Korean Society of Forest Science
• /
• v.97 no.3
• /
• pp.332-339
• /
• 2008

This study was carried out to develop the score table for prediction of landslide hazard in Gyeongsangbuk-Do province. It was studied to 172 places landslided in 23 cities and counties of Gyeongsangbuk-Do province. An analyze of the score table for landslide hazard was carried out through the multiple statistics of quantification method (I) by the computer. Factors effected to landslide occurrence quantity were shown in order of slope position, slope length, bedrock, aspect, forest age, slope form and slope. As results of the development of score table for prediction of landslide hazard in Gyeongsangbuk-Do province, total score range was divided that 107 under is stable area (IV class), 107~176 is area with little susceptibility to landslide (III class), 177~246 is area with moderate susceptibility to landslide (II class), above 247 area with severe susceptibility to landslide (I class).

• Economic and Environmental Geology
• /
• v.46 no.4
• /
• pp.301-312
• /
• 2013

Landslides are caused by complex interaction among a large number of interrelated factors such as topography, geology, forest and soils. In this study, a comparative study was carried out using fuzzy relationship method and artificial neural network to evaluate landslide susceptibility. For landslide susceptibility mapping, maps of the landslide occurrence locations, slope angle, aspect, curvature, lithology, soil drainage, soil depth, soil texture, forest type, forest age, forest diameter and forest density were constructed from the spatial data sets. In fuzzy relation analysis, the membership values for each category of thematic layers have been determined using the cosine amplitude method. Then the integration of different thematic layers to produce landslide susceptibility map was performed by Cartesian product operation. In artificial neural network analysis, the relative weight values for causative factors were determined by back propagation algorithm. Landslide susceptibility maps prepared by two approaches were validated by ROC(Receiver Operating Characteristic) curve and AUC(Area Under the Curve). Based on the validation results, both approaches show excellent performance to predict the landslide susceptibility but the performance of the artificial neural network was superior in this study area.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.37 no.5
• /
• pp.927-940
• /
• 2017

Landslide has been a major disaster in Indonesia, and recent climate change and indiscriminate urban development around the mountains have increased landslide risks. Java Island, Indonesia, where more than half of Indonesia's population lives, is experiencing a great deal of damage due to frequent landslides. However, even in such a dangerous situation, the number of inhabitants residing in the landslide-prone area increases year by year, and it is necessary to develop a technique for analyzing landslide-hazardous and vulnerable areas. In this regard, this study aims to evaluate landslide susceptibility of Java, an island of Indonesia, by using GIS-based spatial prediction models. We constructed the geospatial database such as landslide locations, topography, hydrology, soil type, and land cover over the study area and created spatial prediction models by applying Weight of Evidence (WoE), decision trees algorithm and artificial neural network. The three models showed prediction accuracy of 66.95%, 67.04%, and 69.67%, respectively. The results of the study are expected to be useful for prevention of landslide damage for the future and landslide disaster management policies in Indonesia.

• Journal of the Korean Geosynthetics Society
• /
• v.16 no.4
• /
• pp.129-137
• /
• 2017

In this study, the prediction of landslide probability was performed to the study area located in ${\bigcirc}{\bigcirc}$ area of Muan-gun, Jeonnam Province around Honam railway using the computer program SHAPP ver 1.0 developed by a decision tree model. The soil samples were collected at total 8 points, and soil tests were performed to measure soil properties. The thematic maps of soil properties such as coefficient of permeability and void ratio were made on the basis of soil test results. The slope angle analysis of topography was performed using a digital map. As the prediction result of landslide probability, 435 cells among total 15,552 cells were predicted to be in the event of landslides. Therefore, the predicted area of occurring landslides may be $43,500m^2$ because the analyzed cell size was $10m{\times}10m$.

• Economic and Environmental Geology
• /
• v.52 no.2
• /
• pp.199-212
• /
• 2019

In the machine learning techniques, the sampling strategy of the training data affects a performance of the prediction model such as generalizing ability as well as prediction accuracy. Especially, in landslide susceptibility analysis, the data sampling procedure is the essential step for setting the training data because the number of non-landslide points is much bigger than the number of landslide points. However, the previous researches did not consider the various sampling methods for the training data. That is, the previous studies selected the training data randomly. Therefore, in this study the authors proposed several different sampling methods and assessed the effect of the sampling strategies of the training data in landslide susceptibility analysis. For that, total six different scenarios were set up based on the sampling strategies of landslide points and non-landslide points. Then Random Forest technique was trained on the basis of six different scenarios and the attribute importance for each input variable was evaluated. Subsequently, the landslide susceptibility maps were produced using the input variables and their attribute importances. In the analysis results, the AUC values of the landslide susceptibility maps, obtained from six different sampling strategies, showed high prediction rates, ranges from 70 % to 80 %. It means that the Random Forest technique shows appropriate predictive performance and the attribute importance for the input variables obtained from Random Forest can be used as the weight of landslide conditioning factors in the susceptibility analysis. In addition, the analysis results obtained using specific sampling strategies for training data show higher prediction accuracy than the analysis results using the previous random sampling method.