Search

Long-runout landslides can be extremely hazardous and unpredictable. The prediction of long-runout landslides is challenging because it is unclear what factors control their mobility, and it is unclear how to best measure their runout given that they traverse changing topography. In this paper, we document the development and statistical evaluation of a Landslide Runout Score (LRS) system to predict short, medium, and long runout using a new mobility measure, the unitless Runout Number L/A1/2 (where L is landslide length and A is landslide area). The Runout Number has previously been correlated to three geomorphological factors (planimetric curvature, sand content, and upslope contributing area normalized to landslide area), so the LRS uses these factors as inputs. These factors are readily calculated using geographic information systems (GIS) and publicly available data sources. The LRS system predicts runout categories (short, medium, and long) with one-vs.-all accuracies (i.e., percentage of correct predictions within a given class) of 75, 58, and 72%, respectively, for a total weighted accuracy of approximately 65%. The results of this work are summarized in a worksheet that can be used by geologists and engineers to develop preliminary predictions of landslide runout behavior, which can be incorporated into or used alongside regional-scale landslide hazard assessments. [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]

Mountain areas in the southern western corner of the Kingdom of Saudi Arabia frequently suffer from various types of landslides due to rain storms and anthropogenic activities. To resolve the problem related to landslides, landslide susceptibility map is important as a quick and safe mitigation measure and to help making strategic planning by identifying the most vulnerable areas. This paper summarizes findings of landslide susceptibility analysis at Al-Hasher area, Jizan, KSA, using two statistical models: frequency ratio and index of entropy models with the aid of GIS tools and remote sensing data. The landslide locations (inventory map) were identified in the study area using historical records, interpretation of high-resolution satellite images that include Geo-Eye in 2.5 m and Quickbird in 0.6m resolution, topographic maps of 1:10,000 scale, and multiple field investigations. A total of 207 landslides (80% out of 257 detected landslides) were randomly selected for model training, and the remaining 50 landslides (19%) were used for the model validation. Ten landslide conditioning factors including slope angle, slope-aspect, altitude, curvature, lithology, distance to lineaments, normalized difference vegetation index (NDVI), distance to roads, precipitation, and distance to streams, were extracted from spatial database. Using these conditioning factors and landslide locations, landslide susceptibility and weights of each factor were analyzed by using frequency ratio and index of entropy models. Our findings showed that the existing landslides of high and very high susceptibility classes cover nearly 80.4% and 79.1% of the susceptibility maps produced by frequency ratio and index of entropy models respectively. For verification, receiver operating characteristic (ROC) curves were drawn and the areas under the curve (AUC) were calculated for success and prediction rates. For success rate the results revealed that for the index of entropy model (AUC = 77.9%) is slightly lower than frequency ratio model (AUC = 78.8%). For the prediction rate, it was found that the index of entropy model (AUC = 74.9%) is slightly lower than the frequency ratio model (AUC = 76.7%). The landslide susceptibility maps produced from this study could help decision makers, planners, engineers, and urban areas developers to make suitable decisions. [ABSTRACT FROM AUTHOR]

Landslide susceptibility (LS) assessment by indirect approaches presents some limitations due to (1) the tendency to simplify the environmental factors (i.e., variables) and (2) the assumptions that landslides occur under the same combination of variables for a study site. Recently, some authors have discussed the interest to introduce expert knowledge in the indirect approaches in order to improve the quality of indirect LS maps. However, if the results are reliable, the procedures used seem fastidious and a very good knowledge of the study site is essential. The objectives of this paper are to discuss a methodology to introduce the expert knowledge in the indirect mapping process. After the definition of the expert rules associated to three landslide types, several indirect LS maps are produced by two indirect exploratory approaches, based on fuzzy set theory and on a modification of a bivariate method called expert weight of evidence. Then, the indirect LS maps are confronted to a landslide inventory and a LS map produced by a direct approach. The analyses indicate that the methodology used to introduce the expert rules in the mapping process increases the predictive power of indirect LS map. Finally, some indications about advantages and drawbacks of each approach are given to help the geoscientist to introduce his expert knowledge in the landslide susceptibility mapping process. [ABSTRACT FROM AUTHOR]

In recent years, increasing efforts have been made to predict the time, location, and magnitude of future landslides. This study explores the potential application of four state-of-the-art data mining models (logistic regression, random forest, support vector machine, and Naïve Bayes tree) for the spatially explicit prediction of landslide susceptibility across a landslide-prone landscape in the Zagros Mountains, Iran. Fifteen conditioning factors and 272 historical landslide events were used to develop a geospatial database for the study area. A two-step factor analysis procedure based on the multicollinearity analysis and the Gain Ratio technique was performed to measure the predictive utility of the factors and to quantify their contribution to landslide occurrences across the study region. Once the models were successfully trained and validated using several performance metrics (i.e., ROC-AUC, sensitivity, specificity, accuracy, RMSE, and Kappa), they were applied to the entire study region to generate distribution maps of landslide susceptibilities. Overall, the random forest model demonstrated the highest training performance (AUC = 0.971; accuracy = 99%; RMSE = 0.120) and ability to predict future landslides (AUC = 0.942; accuracy =87%; RMSE = 0.312), followed by the support vector machine, Naïve Bayes tree, and logistic regression models. The Wilcoxon signed-rank test further proved the superiority of the random forest model for mapping landslide susceptibility in the Zagros region. The insights obtained from this research could be useful for the spatially explicit assessment of landslide-prone landscapes and obtaining a better understanding of the capability of different predictive models. [ABSTRACT FROM AUTHOR]

The main objective of this study is to produce a landslide susceptibility map for the Lao Cai area (Vietnam) using a new hybrid intelligent method based on least squares support vector machines (LSSVM) and artificial bee colony (ABC) optimization, namely LSSVM-BC. LSSVM and ABC are state-of-the-art soft computing techniques that have been rarely utilized in landslide susceptibility assessment. LSSVM is adopted to develop landslide prediction model whereas ABC was used to optimize the prediction model by identifying an appropriate set of the LSSVM hyper-parameters. To establish the hybrid intelligent method, a GIS database with ten landslide-influencing factors and 340 landslide locations that occurred mainly during the last 20-years was constructed. These historical landslide locations were collected from the existing inventories that sourced from (i) five landslide projects carried out in this study areas before and (ii) interpretations of SPOT satellite images with resolution of 2.5 m. The study area was geographically split into two different parts, with landslides located in the first part was used for building models whereas the other landslides in the second part was used for the model validation. Performance of the LSSVM-BC model was assessed using the receiver operating characteristic (ROC) curve and area under the curve (AUC). Result shows that the prediction power of the model is good with the area under the curve (AUC) = 0.900. Experiments have pointed out the prediction power of the LSSVM-BC is better than that obtained from the popular support vector machines. Therefore, the proposed model is a promising tool for spatial prediction of landslides at the study area. The landslide susceptibility map is useful for landuse planning for the Lao Cai area. [ABSTRACT FROM AUTHOR]