• 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.

• The Journal of Engineering Geology
• /
• v.32 no.1
• /
• pp.27-40
• /
• 2022

Every year landslides cause serious casualties and property damages around the world. As the accurate prediction of landslides is important to reduce the fatalities and economic losses, various approaches have been developed to predict them. Prediction methods can be divided into landslide susceptibility analysis, landslide hazard analysis and landslide risk analysis according to the type of the conditioning factors, the predicted level of the landslide dangers, and whether the expected consequence cased by landslides were considered. Landslide susceptibility analyses are mainly based on the available landslide data and consequently, they predict the likelihood of landslide occurrence by considering factors that can induce landslides and analyzing the spatial distribution of these factors. Various qualitative and quantitative analysis techniques have been applied to landslide susceptibility analysis. Recently, quantitative susceptibility analyses have predominantly employed the physically based model due to high predictive capacity. This is because the physically based approaches use physical slope model to analyze slope stability regardless of prior landslide occurrence. This approach can also reproduce the physical processes governing landslide occurrence. This review examines physically based landslide susceptibility analysis approaches.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2008.06a
• /
• pp.301-305
• /
• 2008

강원도 지역은 대부분의 지형이 산지로 이루어져 있고, 최근 심각해지고 있는 기후 변화로 인해 집중호우가 잦아지면서 이로 인한 산사태 피해 또한 증가하고 있는 상황이다. 하지만 기존에 이루어져왔던 직접 측량 방식은 많은 시간과 인력이 소모되고, 접근성의 제약으로 인해 곳곳에서 발생하는 모든 산사태를 체계적으로 감지하기에는 무리가 따른다. 따라서 효율적인 산사태 감지와 신속한 대처를 위해 최근 인공위성을 이용한 원격 탐측이 주목을 받고 있으며, 특히 고해상도 영상 레이더(Synthetic Aperture Radar, SAR)는 태양광의 유무나 대기 조건에 상관없이 상시 관측이 가능하다는 장점으로 인해 그 수요가 점점 늘어나고 있는 추세이다. 본 연구에서는 산사태가 집중되는 지역인 강원도 강릉 부근(N $37^{\circ}.30'{\sim}38{\circ}.10'$, E $128^{\circ}.05'{\sim}129^{\circ}.00'$)을 대상으로 SAR 영상 처리 기법 중 하나인 간섭기법(Interferometric SAR, InSAR)를 통해 생성되는 coherence 영상을 분석하여 93년 7월 27일과 동년 9월 9일 사이에 발생한 산사태 피해 지역을 추정하였다.

• The Journal of Engineering Geology
• /
• v.28 no.4
• /
• pp.687-699
• /
• 2018

Understanding rain infiltration into the ground is an important feature of landslide risk evaluation. In this study, a landslide risk index for the study area is suggested, wherein the result of the landslide risk evaluation, based on the factor of safety (FS), is used. The landslide risk index is a landslide risk prediction index that utilizes the saturated depth ratio of the ground. Based on the landslide risk result for the study area, it was found that the FS was first to decrease. However, it gradually became convergent over the 50-year rainfall intensity study period, a result that is similar to the relationship between the saturated depth ratio and soil thickness. Moreover, saturated depth was also found to be deeper on gentle slopes than steep slopes. As such, the landslide risk index, based on the Inhu-ri study result, is thus suggested. Additionally, the suggested landslide risk index was compared and analyzed against the rainfall intensity of previous landslide experience. Results thus revealed that almost all landslides that occurred were over 0.7, which is the second grade, based on the landslide risk index.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.139-139
• /
• 2023

산사태는 사면에서 발생하는 대표적인 토사재해이다. 그리고 산사태가 발생하여 사면이 붕괴하였을 때 동반되어 나타나는 토석류는 지형 변화의 중요한 원인으로 간주한다. 산사태와 토석류가 도시나 농촌 등 인구가 밀집된 지역에서 발생할 경우 직접적인 인명 피해와 재산 피해를 발생시키며, 댐이나 저수지가 위치한 유역에서 발생할 경우 댐/저수지에 토사가 유입되어 유효저수량을 감소시킴으로써 시설물의 기능을 저하할 수 있다. 댐/저수지의 지속적인 운영과 관리하고 이러한 피해를 최소화하기 위해서는 수치 모형을 활용하여 현상을 이해하고, 분석하는 것이 필수적이다. 하지만 한국은 국토 70%의 산지에 약 18,000개의 댐과 저수지가 설치되어 있으나, 댐과 저수지 유역에서 발생하는 산사태와 토석류에 대한 연구는 미흡한 실정이다. 이에 본 연구는 댐이나 저수지 유역에서 발생하는 산사태-토석류로 인해 해당 시설물에 발생하는 피해를 집중적으로 분석하고자 수치모형을 활용하였다. 또한 산지에서 발생하는 토사재해의 특성을 반영하고자 식생을 고려하기 위한 분포 시나리오를 구축하여 사면 안정성 및 토석류 유동에 있어서 식생의 영향을 파악하였습니다.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.2
• /
• pp.61-69
• /
• 2009

In mid-July, 2006 the torrential rainfall across Gangwon-do region caused 48 casualties and 1,248 houses submerged, resulting in damages with the restoration costs of 3 trillion and 512.5 billion won. This was because the topographic characteristics of Gangwon-do region for which mountainous areas mostly account increased the effects of landslide. In this study, the landslide region was shot using the PKNU No.4 system immediately after the occurrence of landslide in order to analyze it as objectively, exactly, and rapidly as possible. 1,054 areas with landslide occurrence were extracted by digitizing the shot images through visual reading after orthometric correction using ERDAS 9.1. Using the Arc GIS 9.2, a GIS program, hydrologic, topographic, clinical, geologic, pedologic aspects and characteristics of the landslide region were established in database through overlay analysis of digital map, vegetation map, geologic map, and soil map, and the status and characteristics of the occurrence of the landslide were analyzed.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.4
• /
• pp.17-24
• /
• 2015

This study utilized high quality three-dimensional geospatial information produced by high-resolution Digital Aerial Photograph and Airborne LiDAR data in order to analyse landslides and debris flows induced by the heavy rainfall in Chuncheon area. Also, this study analysed correlation between the established landslide hazard map and the landslide factor effect and reviewed the analysis result of debris flows on the area where landslides with debris flows occurred frequently. Finally;the study proposed ways to renew the established landslide hazard map effectively and utilize the high quality three-dimensional Geospatial information on the landslide risk area.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.26 no.1
• /
• pp.27-39
• /
• 2008

Landslides cause enormous economic losses and casualties. Korea has mountainous regions and heavy slopes in most parts of the land and has consistently built new roads and large-scale housing complexes according to its industrial and urban growth. As a result, the damage from landslides becomes greater every year. In summer, landslides frequently occur due to local torrential rains and storms. It is critical to predict the potential areas of landslides in advance and to take preventive measures to minimize consequences and to protect property and human life. The previous study on landslides mostly focused on identifying the causes of landslides in the areas where they occurred, and on analyzing landslide vulnerability around the areas without considering rainfall conditions. Thus there were not enough evaluations of the direct risk of landslides to human life. In this study, potentially risky areas for landslides were identified using the GIS data in order to evaluate direct risk on farmlands, roads, and artificial structures that were closely connected to human life. A map of landslide risk was made taking into account rainfall conditions, and a land use map was also drawn with satellite images and digital maps. Both maps were used to identify potentially risky areas for landslides.

• Journal of the Korean Geotechnical Society
• /
• v.30 no.6
• /
• pp.61-68
• /
• 2014

The main objective of this study is to evaluate the preliminary landslide hazard based on the identification of geomorphological features, which are believed to be critical values in the initial state of landslides. Two methods, SINMAP and Planarity analyses, are used to simulate those characteristics where landslides are actually located. Results showed that both methods well discriminate geomorphic features between stable and unstable domains in the landslide areas. SINMAP analysis which is the consecutive model considering external factors like infiltration identifies the landslide hazard especially for debris flow type landslides better than plararity analysis focusing on a specific area. This analysis combined with other methods dealing with specific characteristics of geomorphological feature, the accurate landslide hazard will be evaluated.

• Journal of Korean Society for Geospatial Information Science
• /
• v.15 no.3
• /
• pp.69-76
• /
• 2007

Kangwondo area is mountainous and landslide is easily happened easily during the rainy period in summer time. Especially, when there is torrential downpour caused by the unusual weather change, there will be greater possibility to see landslide. Another reason behind landslide is the continuous forest fire in these several years. Since the surface of the earth has been changed by the fire, when rainfall comes, landslide just happens easily. Also, it is reported that landcover condition, excepted rainfall condition, is the most effect for determining landslide susceptibility area. In this study, it is determined a landslide vulnerable area and landcover information is extracted from four satellite image(Landsat TM), about the landslide vulnerable area, which is pictured for each year. And which distribution change is analyzed. also, NDVI picture is made and distribution change of vegetation vitality is analyzed to study that change of landcover have a effect on landslide. As a result, could know that forest and NDVI are decreasing in landslide vulnerable area.