Search

Data driven methods are widely used for the development of Landslide Susceptibility Mapping (LSM). The results of these methods are sensitive to different factors, such as the quality of input data, choice of algorithm, sampling strategies, and data splitting ratios. In this study, five different Machine Learning (ML) algorithms are used for LSM for the Wayanad district in Kerala, India, using two different sampling strategies and nine different train to test ratios in cross validation. The results show that Random Forest (RF), K Nearest Neighbors (KNN), and Support Vector Machine (SVM) algorithms provide better results than Naïve Bayes (NB) and Logistic Regression (LR) for the study area. NB and LR algorithms are less sensitive to the sampling strategy and data splitting, while the performance of the other three algorithms is considerably influenced by the sampling strategy. From the results, both the choice of algorithm and sampling strategy are critical in obtaining the best suited landslide susceptibility map for a region. The accuracies of KNN, RF, and SVM algorithms have increased by 10.51%, 10.02%, and 4.98% with the use of polygon landslide inventory data, while for NB and LR algorithms, the performance was slightly reduced with the use of polygon data. Thus, the sampling strategy and data splitting ratio are less consequential with NB and algorithms, while more data points provide better results for KNN, RF, and SVM algorithms. [ABSTRACT FROM AUTHOR]

Yang, J., 2018. Landslide damage from extreme rainstorm geological accumulation layers within plain river basins. In: Ashraf, M.A. and Chowdhury, A.J.K. (eds.), Coastal Ecosystem Responses to Human and Climatic Changes throughout Asia. To provide the basis of an early warning and detection for landslide disasters and to provide decision support for local governments in disaster prevention and mitigation, this research studied landslide damage assessment from the geological accumulation layer in a plain river basin under conditions of an extreme rainstorm. Based on an analysis of the relationship between landslide geology and rainfall and a damage model of landslides under conditions of extreme rainfall, the landslide damage assessment described in this article used the SINMAP model. The layer of geological accumulation in the Wenchuan–Diexi River section in the upper reaches of the Minjiang River was the study area. Spatial distribution characteristics and the main factors inducing a landslide disaster were studied, and a danger-evaluation model for landslides in that region was established for landslide assessment research and for applying that research. Based on the SINMAP model, the effects of different extreme rainstorm events on the stability of regional slopes were analyzed. The results of a stability evaluation for landslides show that the area has an unstable zone of 4339.8 km2, which accounts for 64.49% of the total area. The unstable area in this evaluation was basically consistent with the distribution of an existing landslide, which shows that the model can adequately evaluate slope stability and landslide potential from the layer of geological accumulation in this area. [ABSTRACT FROM AUTHOR]

The dynamic monitoring of landslides in engineering geology has focused on the correlation among landslide stability, rainwater infiltration, and subsurface hydrogeology. However, the understanding of this complicated correlation is still poor and inadequate. Thus, in this study, we investigated a typical landslide in southwestern China via time-lapse electrical resistivity tomography (TLERT) in November 2013 and August 2014. We studied landslide mechanisms based on the spatiotemporal characteristics of surface water infiltration and flow within the landslide body. Combined with borehole data, inverted resistivity models accurately defined the interface between Quaternary sediments and bedrock. Preferential flow pathways attributed to fracture zones and fissures were also delineated. In addition, we found that surface water permeates through these pathways into the slipping mass and drains away as fissure water in the fractured bedrock, probably causing the weakly weathered layer to gradually soften and erode, eventually leading to a landslide. Clearly, TLERT dynamic monitoring can provide precursory information of critical sliding and can be used in landslide stability analysis and prediction. [ABSTRACT FROM AUTHOR]

This work focuses on the development of a combined statistical-mechanical approach to predict changes in landslide displacement rates from observed changes in rainfall amounts. The forecasting tool FLAME (Forecasting Landslides Accelerations induced by Meteorological Events) associates (1) a statistical impulse response (IR) model to simulate the changes in landslide rates by computing a transfer function between the input signal (e.g. rainfall) and the output signal (e.g. displacement) and (2) a simple 1D mechanical (MA) model (e.g. viscoplastic rheology) to take into account changes in pore water pressure. The models have been applied to forecast the displacement rates at the Super-Sauze landslide (South East France). The performance of different combinations of models (IR model alone, MA model alone and a combination of the IR and MA models) is evaluated against observed changes in pore water pressures and displacement rates at the study site. Results indicate that the three models are able to reproduce the displacement pattern in the general kinematic regime (succession of acceleration and deceleration phases); conversely, extreme kinematic regimes such as fluidization of part of the landslide mass are not being reproduced. The approach constitutes however a robust tool to predict changes in displacement rates from rainfall or groundwater time series. [ABSTRACT FROM AUTHOR]

Landslides are a common type of mass movement that occur in large numbers and have significant effects on topography. In Turkey, landslides rank second after earthquakes among natural disasters that cause widespread loss of life and property. This study investigates the monitoring of the Taşkent Landslide (Konya, Turkey), which has considerably affected the local topography. The landslide in Taşkent had an adverse effect on transportation to nearby towns and villages. Initial studies were conducted using the global navigation satellite system (GNSS) measurement scale; these studies are currently on-going. Since 2011, studies using terrestrial laser scanning (TLS), which is one of the latest ground-positioning technologies, have also been launched. This article presents the results of the TLS and GNSS approaches used to monitor, characterise and model an active landslide. It includes the comparison and analysis of a point cloud, three-dimensional model, profile and change in volume mass acquired in different time periods in one landslide area. The point cloud data, obtained from two different time periods using the TLS technique, and supported by the results of GNSS analysis, were analysed via different methods. The TLS revealed that up to 4 m of material was displaced at the site of the landslide, and it revealed detailed information about how the landslide occurred. The volume of the mass that moved as a result of the landslide was determined, and movements that occurred in the geological structure were monitored. These studies showed that the TLS point cloud method can be effectively used in the monitoring of landslides that occur in areas that are large, difficult to reach, rugged or mountainous. [ABSTRACT FROM AUTHOR]

An ensemble algorithm of data mining decision tree (DT)-based CHi-squared Automatic Interaction Detection (CHAID) is widely used for prediction analysis in variety of applications. CHAID as a multivariate method has an automatic classification capacity to analyze large numbers of landslide conditioning factors. Moreover, it results two or more nodes for each independent variable, where every node contains numbers of presence or absence of landslides (dependent variable). Other DT methods such as Quick, Unbiased, Efficient Statistic Tree (QUEST) and Classification and Regression Trees (CRT) are not able to produce multi branches based tree. Thus, the main objective of this paper is to use CHAID method to perform the best classification fit for each conditioning factors, then, combined it with logistic regression (LR) to find the corresponding coefficients of best fitting function that assess the optimal terminal nodes. In the first step, a landslide inventory map with 296 landslide locations were extracted from various sources over the Pohang-Kyeong Joo catchment (South Korea). Then, the inventory was randomly split into two datasets, 70 % was used for training the models, and the remaining 30 % was used for validation purpose. Thirteen landslide conditioning factors were used for the susceptibility modeling. Then, CHAID was applied and revealed that some conditioning factors such as altitude, soil drain, soil texture and TWI, as terminal nodes and reflected the best classification fit. Then, a proposed ensemble technique was applied and the interpretations of the coefficients showed that the relationship between the decision tree branch nodes distance from drain, soil drain, and TWI, respectively, leads to better consequences assessment of landslides in the current study area. The validation results showed that both success and prediction rates, 75 and 79 %, respectively. This study proved the efficiency and reliability of ensemble DT and LR model in landslide susceptibility mapping. [ABSTRACT FROM AUTHOR]

Rainfall-induced landslides form an important natural threat in Vietnam. The purpose of this study is to explore regional landslide susceptibility mapping in the mountainous district of A Luoi in Thua Thien Hue Province, where data on the occurrence and causes of landslides are very limited. Three methods are applied to examine landslide susceptibility: statistical index, logistic regression and certainty factor. Nine causative factors are considered: elevation, slope, geological strata, fault density, geomorphic landforms, weathering crust, land use, distance to rivers and annual precipitation. The reliability of the landslide susceptibility maps is evaluated by a receiver operating characteristic curve and the area under the curve is used to quantify and compare the prediction accuracy of the models. The certainty factor model performs best. This model is optimized by maximizing the difference between the true positive rate and the false positive rate. The optimal model correctly identifies 84% of the observed landslides. The results are verified with a validation test, whereby the model is calibrated with 75% randomly selected observed landslides, while the remaining 25% of the observed landslides are used for validation. The validation test correctly identifies 81% of the observed landslides in the training set and 73% of the observed landslides in the validation set. [ABSTRACT FROM AUTHOR]