• The Journal of Engineering Geology
• /
• v.30 no.3
• /
• pp.315-325
• /
• 2020

This study was performed to develop a model to predict landslides and determine the variable importance of landslides susceptibility factors based on the probabilistic prediction of landslides occurring on slopes along the road. Field survey data of 30,615 slopes from 2007 to 2020 in Korea were analyzed to develop a landslide prediction model. Of the total 131 variable factors, 17 topographic factors and 114 geological factors (including 89 bedrocks) were used to predict landslides. Automated machine learning (AutoML) was used to classify landslides and non-landslides. The verification results revealed that the best model, an extremely randomized tree (XRT) with excellent predictive performance, yielded 83.977% of prediction rates on test data. As a result of the analysis to determine the variable importance of the landslide susceptibility factors, it was composed of 10 topographic factors and 9 geological factors, which was presented as a percentage for each factor. This model was evaluated probabilistically and quantitatively for the likelihood of landslide occurrence by deriving the ranking of variable importance using only on-site survey data. It is considered that this model can provide a reliable basis for slope safety assessment through field surveys to decision-makers in the future.

• The Journal of Engineering Geology
• /
• v.32 no.4
• /
• pp.585-596
• /
• 2022

This study determined major factors influencing landslide occurrence along a forest road near Sangsan village, Sancheok-myeon, Chungju-si, Chungcheongbuk-do, South Korea. Within a 2 km radius of the study area, landslides occur intensively during periods of heavy rainfall (August 2020). This makes study of the area advantageous, as it allows examination of the influence of only geological and tomographic factors while excluding the effects of rainfall and vegetation. Data for 82 locations (37 experiencing landslides and 45 not) were obtained from geological surveys, laboratory tests, and geo-spatial analysis. After some data preprocessing (e.g., error filtering, minimum-maximum normalization, and multicollinearity), structural equation model (SEM) and logistic regression (LR) analyses were conducted. These showed the regolith thickness, porosity, and saturated unit weight to be the factors most influential of landslide risk in the study area. The sums of the influence magnitudes of these factors are 71% in SEM and 83% in LR.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2007.06a
• /
• pp.413-423
• /
• 2007

우리나라의 자연재해는 기상학적 자연현상에 의해 주로 발생되고 있으며 그 발생원인은 태풍, 호우, 폭풍, 폭풍우, 재설, 폭풍성 우박, 해일 및 기타(낙뢰, 돌풍, 설해, 결빙, 지진 등을 포함)로 구분되며 이중 발생빈도가 가장 높은 것은 강우에 의한 재해로 전체 재해발생 원인 중 약 80%로 대부분을 차지하고 있다. 특히 사면붕괴와 관련된 자연재해(산사태, 옹벽붕괴, 매몰 등)는 최근 국지성 집중호우를 포함하여 호우의 집중 강도가 높아지는 등 기상학적 원인에 의해 매년 발생하고 있다. 따라서 우리나라에서 발생되는 자연재해와 관련한 사면붕괴의 특성을 강우특성에 따라 조사 분석할 필요가 있으며 이에 적합한 대책들이 더욱 필요하다. 이 연구에서는 산사태 유발인자와 강우조건을 고려하여 산사태 잠재가능성을 평가하고 산사태 취약지역을 분석하여 지역적인 강우특성을 고려한 산사태 가능성을 평가하였다.

• Journal of Korean Society of Forest Science
• /
• v.97 no.3
• /
• pp.326-331
• /
• 2008

This study was carried out to analyze the characteristics of each factors by using the quantification theory(I) for prediction of landslide hazard area. The results obtained from this study were summarized as follows; The stepwise regression analysis between landslide sediment ($m^3$ ) and environmental factors, factors affecting landslide sediment ($m^3$ ) were high in order of mixed (forest type), < 15 cm(soil depth), 801~1,200 m (altitude), $31{\sim}40^{\circ}$ (slope gradient), 46 cm < (soil depth), 1,201 m < (altitude) and s(aspect). According to the range, it was shown in order of soil depth (0.3784), altitude (0.2876), forest type (0.2409), slope gradient (0.1728) and aspect (0.1681). The prediction of landslide hazard area was estimated by score table of each category. The extent of prediction score was 0 to 1.2478, and middle score was 0.6239. Class I was over 1.1720, class II was 0.7543 to 0.1719, class III was 0.4989 to 0.7542 and class IV was below 0.4988.

• Journal of Korean Society of Forest Science
• /
• v.103 no.2
• /
• pp.233-239
• /
• 2014

This study was carried out to propose the criteria for the prediction of landslide occurrence through analysis the influence of each factor by using the quantification theory. The results obtained from this study are summarized as follows. From a stepwise regression analysis between the landslide area($m^2$) and environmental factors, the factors strongly affecting the landslide sediment($m^2$) were the Parents rock (igneous), cross slope(complex), coniferous forests (forest type) and slope gradient ($21{\sim}30^{\circ}$). According to the range, it was shown in order of Cross slope (0.2922), Parents rock (0.2691), Forest type (0.2631) and Slope gradient (0.2312). The range of prediction score of landslide occurrence has been distributed between score 0 and score 1.0556, the median value was score 0.5278. The prediction for class I was over 0.7818, for class II was 0.5279 to 0.7917, for class III 0.2694 to 0.5278 and for class IV was below 0.2693. The prediction on landslide occurrence appeared relatively high accuracy rate as 72% for class I and II. Therefore, this score table for landslide will be very useful for judgement of dangerous slope.

• The Journal of Engineering Geology
• /
• v.33 no.4
• /
• pp.673-687
• /
• 2023

Exploratory data analysis is the process of observing and understanding data collected from various sources to identify their distributions and correlations through their structures and characterization. This process can be used to identify correlations among conditioning factors and select the most effective factors for analysis. This can help the assessment of landslide susceptibility, because landslides are usually triggered by multiple factors, and the impacts of these factors vary by region. This study compared two stages of exploratory data analysis to examine the impact of the data exploration procedure on the landslide prediction model's performance with respect to factor selection. Deep-learning-based landslide susceptibility analysis used either a combinations of selected factors or all 23 factors. During the data exploration phase, we used a Pearson correlation coefficient heat map and a histogram of random forest feature importance. We then assessed the accuracy of our deep-learning-based analysis of landslide susceptibility using a confusion matrix. Finally, a landslide susceptibility map was generated using the landslide susceptibility index derived from the proposed analysis. The analysis revealed that using all 23 factors resulted in low accuracy (55.90%), but using the 13 factors selected in one step of exploration improved the accuracy to 81.25%. This was further improved to 92.80% using only the nine conditioning factors selected during both steps of the data exploration. Therefore, exploratory data analysis selected the conditioning factors most suitable for landslide susceptibility analysis and thereby improving the performance of the analysis.

• 한국방재학회:학술대회논문집
• /
• 2008.02a
• /
• pp.353-356
• /
• 2008

최근 대규모 산림재해로 인한 산림환경 훼손 및 산림 농가의 피해는 물론 산림생태계에도 나쁜 영향을 미치고 있으며 이는 사회적으로 매우 민감한 환경문제로서 국민의 주요 관심사가 되고 있다. 본 연구에서는 울진군 전체를 대상으로 GIS 및 RS 기법을 이용하여 다양한 산사태 관련 인자들을 추출 하여 이를 기반으로 GIS 중첩 및 가중치 분석을 통하여 울진군의 산사태 발생 가능 위험지역의 분포도를 작성하고자 한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.705-705
• /
• 2012

우리나라는 최근 여름철 태풍 및 집중호우로 인해 많은 토석류 및 산사태가 발생하고 있다. 작년 7월에도 집중호우로 인해 서울시 우면산 일대와 강원도 춘천에 많은 인적 물적 피해를 입었다. 해마다 반복되는 토석류나 산사태의 위험을 감소시키기 위해서는 보다 정확한 위험지역 예측모델을 필요로 한다. 본 연구는 토석류 및 산사태의 위험과 취약지역을 예측하기 위하여 GIS기반의 Weight of Evidence 기법을 적용하여 위험지역을 분석 하고자 한다. 2006년 태풍 에위니아에 의해 많은 토석류 피해를 입은 강원도 인제군 가리산일대를 대상으로 하였으며 토석류 및 산사태 위치 자료는 2005년, 2006년 토석류 발생 전후 항공사진의 중첩분석을 토대로 발생 지역을 추출하였다. 토석류 및 산사태발생에 영향을 미치는 지형, 지질, 토양, 수문, 임상 등의 인자들은 GIS를 이용하여 DB로 구축하였다. 베이시안 확률기법(Bayesian Method)에 기반 하여 구축된 DB와 결합하여 각각의 인자의 가중 값 W+, W-를 계산하여 상관관계를 분석하고 Weight of Evidence 기법을 적용하여 위험지역을 정량적으로 평가하고자 한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.628-628
• /
• 2015

유역의 형상은 강우, 산사태 등과 같은 지배적인 침식작용과 더불어 지형 지질학적 요인들에 의해 결정되어 진다. 그러므로 유역형상에 대한 공간특성 분석을 위해서는 지형학적 요인과 다양한 침식작용에 대한 분석이 필요하다. 국내 외 많은 연구결과에 의하면 지형학적 인자에 의한 침식 형태는 국부경사와 집수면적의 크기에 의해 다양한 구간으로 나뉘며, 그 특성에 따라 지표침식, 세굴, 산사태 등으로 구분되는 것으로 연구된 바 있다. 일례로 유역 내 세굴과 관련된 지배인자는 집수면적보다는 국부경사에 반대로 지표침식, 산사태는 국부경사보다는 집수면적의 크기에 따라 영향을 받는다. 따라서 지형학적 인자(국부경사, 집수면적)의 임계치(threshold) 산출을 통해 침식특성(불안정지역)을 검토할 수 있으며, 이에 대한 물리적 검증은 여러 연구를 통해 물질플럭스(유량, 에너지)에 대한 Power Law로써 검증된바 있다. 본 연구에서는 이러한 지형학적 침식특성 분석을 위하여 2006년 집중호우에 의해 광역적 산사태가 발생한 강원도 평창군 진부면 일대의 $10m{\times}10m$ DEM로부터 국부경사, 집수면적을 산출하고 경사-면적한계곡선(Slope-Area Threshold Curve, SATC), 배수면적 확률분포곡선(Probability distribution of Drain Areas Curve, PDAC), 에너지지수 확률분포곡선(Probability distribution of Energy Index Curve, PEIC)를 실제 산사태지점과 중첩하여 도시하였다. 그 결과, 특정 임계구간(Threshold Area, Unstable area, 2~3권역)내에서 산사태 발생지점이 분포하는 것으로 분석되었다. 이를 통해 지형학적 인자만을 고려하여 미계측 유역에 대한 잠재적 불안정지역의 판별이 가능할 것으로 판단되며, 추후 광역적 사면안정해석에 적용 가능할 것으로 판단된다.

• 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.