Search

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]

The frequency of landslides and their magnitude are increasing worldwide due to global climate change, causing damage to people and infrastructure. Therefore, landslide risk assessment is urgent. This study developed a novel landslide risk assessment framework to analyze landslide risk on Mt. Umyeon, Republic of Korea. The proposed framework comprises four main procedures: (1) landslide hazard analyses using an ensemble statistical and physical model, (2) analyses of physical vulnerability from a vulnerability curve, (3) analyses of physical vulnerability from a semi-quantitative approach, and (4) risk index calculation using the results from the previous steps via a proposed equation which combines the quantitative and semi-quantitative approaches. To confirm the reliability of the framework, the results of each step were compared to a real landslide event that occurred in July 2011 on Mt. Umyeon. The framework successfully assessed landslide risk in this area. In addition, the landslide risk for different landslide return periods was calculated. This new framework for landslide risk assessment supports reliable decision-making for landslide risk assessment and management. [ABSTRACT FROM AUTHOR]

The intensity-duration (I-D) threshold is considered an effective indicator for landslides triggered by short-term high-intensity rainfall and long-term low-intensity rainfall. However, previous studies have not considered the influence of antecedent rainfall. Herein, we analyzed hourly rainfall data for 613 shallow landslides that occurred from 1963 to 2018 in South Korea to obtain rainfall thresholds and investigated the effect of antecedent rainfall conditions on threshold variations. The I-D and IMAP-D, which is normalized by dividing I by mean annual precipitation (MAP), thresholds were determined to be I = 10.40D–0.31 and IMAP = 0.006D–0.26 (4 ≤ D (h) ≤ 84), respectively, at the 2nd percentile level through quantile regression analysis. These thresholds were lower than other local thresholds (i.e., excluding global and regional thresholds), suggesting that the southern region of the Korean Peninsula was more susceptible to rainfall-induced landslides. Although the effective length of antecedent rainfall was not presented herein, the I-D threshold of landslide-triggering rainfall was confirmed to be low for the absolute and/or calibrated antecedent rainfalls greater than event cumulative rainfall at 5, 7, 10, and 20 days prior to the event. Such differences in I can be greater at shorter durations; however, these differences gradually decrease as D increases, suggesting that they lose their effect as soil water content increases. The results of the current study can improve the understanding of the effect of antecedent rainfall conditions on landslide occurrence and should be further tested with respect to the hydrologic response of hillslopes by considering regional climate and local site conditions. [ABSTRACT FROM AUTHOR]