Spatial Prediction Of Landslides Research Articles

Space-time modeling of landslide size by combining static, dynamic, and unobserved spatiotemporal factors

Landslide spatial prediction using data-driven models has predominantly concentrated on predicting where landslides may occur. Nevertheless, few researchers have turned to jointly modeling how large and when landslides will be for a given terrain unit. This study proposes a data-driven model capable of estimating how large landslides may be, for the entire Taiwan main island in a fourteen-year time window. To address this task, we implement a space–time generalized additive model to fit the complex relationships between environmental factors and landslide size. In addition to incorporating static and dynamic covariates into the modeling process, the model takes into account spatial and temporal interactions to elucidate the spatiotemporal variations affecting landslide size. To test the effectiveness of the model, we employ a comprehensive set of cross-validation (CV) procedures, includes a randomized 10fold-CV, a spatially constrained CV, a temporal leave-one-year-out CV, and a spatio-temporal CV. The experimental results demonstrate that the space–time model delivers acceptable and interpretable prediction outcomes, demonstrating the ability to predict landslide area for a given slope unit within a specified time period. We believe that the space–time landslide modeling will lay the foundation for landslide community to analyze landslide characteristic within a dynamic context. Furthermore, given its inherent spatio-temporal nature, we anticipate that this approach may pave the way for simulation studies exploring diverse climate scenarios.

Landslide spatial prediction using cluster analysis

Temporal clustering is an intrinsic nature of landslide occurrences, therefore it should be considered in data-driven landslide spatial prediction (i.e., susceptibility assessment). However, it remains problematic regarding how to determine landslide temporal clusters and how to integrate susceptibility maps derived from different landslide temporal clusters. In this paper, a general framework of landslide spatial prediction model considering the temporal clustering of landslides is proposed. This novel framework first assesses landslide susceptibility separately based on each landslide temporal cluster identified by spatiotemporal clustering analysis and then integrates separate assessments by weighted averaging. In a case study, this general framework is implemented using the stacking network landslide susceptibility assessment method and used in the landslide spatial prediction of the Sanming City and Wenchuan seismic areas. The results show that the proposed framework outperformed traditional susceptibility models that do not consider landslide temporal clustering, and the integration of susceptibility models based on all landslide temporal clusters will promote the performance of landslide spatial prediction because levels of knowledge in long-term spatiotemporal landslide activities are considered. This novel general framework highlights the benefit of considering landslide temporal clustering in landslide spatial prediction and can provide better support for landslide risk management.

Effect of time-variant rainfall on landslide susceptibility: A case study in Quang Ngai Province, Vietnam

Rainfall is a triggering factor that causes landslides, especially in the regions where landslides often occur after consecutive days of heavy rainfall. Most previous studies only used a specific rainfall map for landslide susceptibility assessment. However, this approach was unreasonable because rainfall is a time-variant data. This study uses the time series data of 1-day, 3-day, 5-day, and 7-day maximum precipitation from 2016 to 2020 in the mountainous area of Quang Ngai province for landslide susceptibility assessment. These data and other influencing factors were used to develop landslide spatial prediction models using the Extreme Gradient Boosting method. The prediction model's performance was assessed using the statistical index and receiver operating characteristic curve methods. The testing results of 4 cases using consecutive days of maximum rainfall data demonstrated excellent performance. Of these, the model with a 3-day maximum rainfall with ACC = 0.813, kappa = 0.625, SST = 0.872, SPF = 0.754, and AUC = 0.895 had the best performance. In addition, these results were compared to the previous approach that used average annual rainfall. The validation result indicates that the cases using a time series of maximum precipitation (with AUC of approximately 0.9) outperform the cases with average annual rainfall (AUC=0.838). Finally, the model using 3-day maximum rainfall is then used for landslide spatial prediction mapping. These maps provide spatial prediction and assess landslide susceptibility corresponding to rainfall frequencies.

Landslide spatial prediction utilizing fuzzy unordered rules induction ensemble models: a case study in Thai Nguyen, Vietnam

Landslide spatial prediction utilizing fuzzy unordered rules induction ensemble models: a case study in Thai Nguyen, Vietnam

Data-Driven Landslide Spatial Prediction and Deformation Monitoring: A Case Study of Shiyan City, China

Landslide susceptibility mapping (LSM) is significant for landslide risk assessment. However, there remains no consensus on which method is optimal for LSM. This study implements a dynamic approach to landslide hazard mapping by integrating spatio-temporal probability analysis with time-varying ground deformation velocity derived from the MT-InSAR (Multi-Temporal InSAR) method. Reliable landslide susceptibility maps (LSMs) can inform landslide risk managers and government officials. First, sixteen factors were selected to construct a causal factor system for LSM. Next, Pearson correlation analysis, multicollinearity analysis, information gain ratio, and GeoDetector methods were applied to remove the least important factors of STI, plan curvature, TRI, and slope length. Subsequently, information quantity (IQ), logistic regression (LR), frequency ratio (FR), artificial neural network (ANN), random forest (RF), support vector machine (SVM), and convolutional neural network (CNN) methods were performed to construct the LSM. The results showed that the distance to a river, slope angle, distance from structure, and engineering geological rock group were the main factors controlling landslide development. A comprehensive set of statistical indicators was employed to evaluate these methods’ effectiveness; sensitivity, F1-measure, and AUC (area under the curve) were calculated and subsequently compared to assess the performance of the methods. Machine learning methods’ training and prediction accuracy were higher than those of statistical methods. The AUC values of the IQ, FR, LR, BP-ANN, RBF-ANN, RF, SVM, and CNN methods were 0.810, 0.854, 0.828, 0.895, 0.916, 0.932, 0.948, and 0.957, respectively. Although the performance order varied for other statistical indicators, overall, the CNN method was the best, while the BP-ANN and RBF-ANN method was the worst among the five examined machine methods. Hence, adopting the CNN approach in this study can enhance LSM accuracy, catering to the needs of planners and government agencies responsible for managing landslide-prone areas and preventing landslide-induced disasters.

Towards landslide space-time forecasting through machine learning: the influence of rainfall parameters and model setting

Landslide susceptibility assessment using machine learning models is a popular and consolidated approach worldwide. The main constraint of susceptibility maps is that they are not adequate for temporal assessments: they are generated from static predisposing factors, allowing only a spatial prediction of landslides. Recently, some methodologies have been proposed to provide spatiotemporal landslides prediction starting from machine learning algorithms (e.g., combining susceptibility maps with rainfall thresholds), but the attempt to obtain a dynamic landslide probability map directly by applying machine learning models is still in the preliminary phase. This work provides a contribution to fix this gap, combining in a Random Forest (RF) algorithm a static indicator of the spatial probability of landslide occurrence (i.e., a classical susceptibility index) and a number of dynamic variables (i.e., seasonality and the rainfall amount cumulated over different reference periods). The RF implementation used in this work allows the calculation of the Out-of-Bag Error and depicts Partial Dependence Plots, two indices that were used to quantify the variables’ importance and to comprehend if the model outcomes are consistent with the triggering mechanism observed in the case of study (Metropolitan City of Florence, Italy). The goal of this research is not to set up a landslide probability map, but to 1) understand how to populate training and test datasets with observations sampled over space and time, 2) assess which rainfall variables are statistically more relevant for the identification of the time and location of landslides, and 3) test the dynamic application of RF in a forecasting model for the spatiotemporal prediction of landslides. The proposed dynamic methodology shows encouraging results, consistent with the actual knowledge of the physical mechanism of the triggering of shallow landslides (mainly influenced by short and intense rainfalls) and identifies some benchmark configurations that represents a promising starting point for future regional-scale applications of machine learning models to dynamic landslide probability assessment and early warning.

Transfer Learning with Attributes for Improving the Landslide Spatial Prediction Performance in Sample-Scarce Area Based on Variational Autoencoder Generative Adversarial Network

Owing to the complexity of obtaining the landslide inventory data, it is a challenge to establish a landslide spatial prediction model with limited labeled samples. This paper proposed a novel strategy, namely transfer learning with attributes (TLAs), to make good use of existing landslide inventory data, a strategy that is based on a variational autoencoder of a generative adversarial network (VAEGAN) for improving the landslide spatial prediction performance in sample-scarce areas. Different from transfer learning (TL), TLAs are pretraining the model with the data reconstructed by VAEGAN, so that the models learn in advance the landslide attributes of sample-scarce areas. Accordingly, a database containing a total of 986 landslides in three study areas with 14 landslide-influencing factors was established, and each of the three models, i.e., convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM) and gated recurrent units (GRUs), was respectively selected as the feature extractor of the VAEGAN to reconstruct the data with attributes and the prediction model to generate the landslide susceptibility maps to investigate and validate the proposed TLA strategy. The experimental results showed that the TLA strategy increased the mean value of evaluators, such as the area under the receiver-operating characteristic (AUROC), F1-score, precision, recall and accuracy by about 2–7% compared with TL, results that indicated that the generated data have the attribute of specific study areas and the effectiveness of TLA strategy in sample-scare areas.

Application of convolutional neural networks based on Bayesian optimization to landslide susceptibility mapping of transmission tower foundation

The stability of tower foundation slopes is an important factor to maintain the operation of a power system. However, it is time-consuming and expensive to evaluate tower foundation slopes one by one due to the large area. The aim of this study is to investigate the performance of CNNs with different architectures and training options for transmission tower foundation landslide spatial prediction (LSP) by Bayesian optimization. Accordingly, fourteen influencing factors related to landslide evaluated by gain ratio technique are considered and 424 historical landslide locations in Luoding and Xinyi Counties (Guangdong Province, China) are randomly divided into 80% for training and 20% for testing the CNNs. The CNN performances are investigated by permutating and combining different numbers of convolutional layers, pooling layers and learning rate strategy. In 59 Bayesian optimized cases, three conclusions are drawn: (a) the CNNs yielded the best result with 3 convolution layers, (b) the CNN without a pooling layer performs best, and (c) a piece-wise decay learning rate strategy yields better performance. Meanwhile, the excellent performance of the CNN obtained by Bayesian optimization (CNNB) has also been validated by comparisons with gravitational search optimization algorithm and other landslide spatial models, which indicates that CNNB can be applied to generate the susceptibility maps for locating transmission tower foundations in high landslide susceptibility zones and reducing the impact of landslides on power supply by taking measures in advance.

Sensitivity evaluation of landslide geological hazards based on Multi-source Remote Sensing Data

The landslide geological disaster caused by heavy rainfall has the characteristics of sudden and rain and slide. Due to the limitations of weather, terrain and other conditions during the disaster period, the traditional manual survey is difficult to carry out at the first time and can not meet the needs of disaster relief. Multi source remote sensing data makes up for the defects of a single sensor. Relying on its advantages of large-scale, long time series and all-weather real-time imaging, using multi-source remote sensing data technology, various evaluation elements in the survey area can be quickly extracted to investigate the sensitivity evaluation of landslide geological hazards. Landslide is one of the natural disasters that cause great losses and casualties in the world. This is a serious geological disaster second only to earthquake, which poses a great threat to people's life, property safety and economic construction. Considering the serious situation of landslide disaster and the huge losses caused by it, effectively investigating and monitoring the distribution and activities of landslide, predicting and evaluating landslide are not only the urgent requirements of the affected areas, but also the mission of landslide scientists, so as to provide a scientific basis for the prevention and management of landslide disaster. This paper will study the landslide sensitivity evaluation, which is the central problem of landslide disaster research. Landslide sensitivity, also known as landslide susceptibility, belongs to the category of landslide spatial prediction. In this paper, GIS, RS, data mining and landslide sensitivity evaluation are combined to solve the problems of low efficiency, low precision, time and labor consumption in the previous landslide sensitivity evaluation, and realize the informatization and scientization of landslide geology.

Comparing the prediction performance of logistic model tree with different ensemble techniques in susceptibility assessments of different landslide types

Susceptibility based on different landslide types has rarely been assessed. Therefore, this paper aims to compare the prediction performance of hybrid approaches by combining the logistic model tree (LMT) with Decorate, random subspace, and rotation forest ensemble techniques (De-LMT, RS-LMT, and RoF-LMT) for susceptibility assessments of different landslide types. The Yuzi River catchment along the eastern margin of the Tibetan Plateau was selected as the study area. In this catchment, 478 rockfalls and 167 landslides were identified, and further partitioned into training (70%) and validation (30%) datasets, respectively. Subsequently, 13 conditioning factors were initially considered, and then 11 and 8, respectively, were selected as the final rockfall and landslide influencing parameters to conduct susceptibility models through multicollinearity analysis and information gain. Finally, the model performances were evaluated by area under the receiver operating characteristic curves (AUC). The obtained results demonstrated that elevation and rainfall were most influential factors for rockfall and landslide. For the rockfall prediction, the De-LMT model achieved the best prediction accuracy with the highest AUC (0.939), followed by the RS-LMT (0.938), RoF-LMT (0.928), and LMT (0.919) models. For the landslide prediction, the De-LMT model (AUC = 0.931) outperformed and outclassed the RoF-LMT (0.926), RS-LMT (0.922), and LMT (0.915) models. Therefore, it is reasoned out that all models exhibited satisfactory performance (AUC > 0.9), and the De-LMT model was superior in rockfall and landslide spatial prediction.

An attention-constrained neural network with overall cognition for landslide spatial prediction

An attention-constrained neural network with overall cognition for landslide spatial prediction

Hybrids of Support Vector Regression with Grey Wolf Optimizer and Firefly Algorithm for Spatial Prediction of Landslide Susceptibility

Landslides are one of the most frequent and important natural disasters in the world. The purpose of this study is to evaluate the landslide susceptibility in Zhenping County using a hybrid of support vector regression (SVR) with grey wolf optimizer (GWO) and firefly algorithm (FA) by frequency ratio (FR) preprocessed. Therefore, a landslide inventory composed of 140 landslides and 16 landslide conditioning factors is compiled as a landslide database. Among these landslides, 70% (98) landslides were randomly selected as the training dataset of the model, and the other landslides (42) were used to verify the model. The 16 landslide conditioning factors include elevation, slope, aspect, plan curvature, profile curvature, distance to faults, distance to rivers, distance to roads, sediment transport index (STI), stream power index (SPI), topographic wetness index (TWI), normalized difference vegetation index (NDVI), landslide, rainfall, soil and lithology. The conditioning factors selection and spatial correlation analysis were carried out by using the correlation attribute evaluation (CAE) method and the frequency ratio (FR) algorithm. The area under the receiver operating characteristic curve (AUROC) and kappa data of the training dataset and validation dataset are used to evaluate the prediction ability and the relationship between the advantages and disadvantages of landslide susceptibility maps. The results show that the SVR-GWO model (AUROC = 0.854) has the best performance in landslide spatial prediction, followed by the SVR-FA (AUROC = 0.838) and SVR models (AUROC = 0.818). The hybrid models of SVR-GWO and SVR-FA improve the performance of the single SVR model, and all three models have good prospects for regional-scale landslide spatial modeling.

A New Integrated Approach for Landslide Data Balancing and Spatial Prediction Based on Generative Adversarial Networks (GAN)

Landslide susceptibility mapping has significantly progressed with improvements in machine learning techniques. However, the inventory/data imbalance (DI) problem remains one of the challenges in this domain. This problem exists as a good quality landslide inventory map, including a complete record of historical data, is difficult or expensive to collect. As such, this can considerably affect one’s ability to obtain a sufficient inventory or representative samples. This research developed a new approach based on generative adversarial networks (GAN) to correct imbalanced landslide datasets. The proposed method was tested at Chukha Dzongkhag, Bhutan, one of the most frequent landslide prone areas in the Himalayan region. The proposed approach was then compared with the standard methods such as the synthetic minority oversampling technique (SMOTE), dense imbalanced sampling, and sparse sampling (i.e., producing non-landslide samples as many as landslide samples). The comparisons were based on five machine learning models, including artificial neural networks (ANN), random forests (RF), decision trees (DT), k-nearest neighbours (kNN), and the support vector machine (SVM). The model evaluation was carried out based on overall accuracy (OA), Kappa Index, F1-score, and area under receiver operating characteristic curves (AUROC). The spatial database was established with a total of 269 landslides and 10 conditioning factors, including altitude, slope, aspect, total curvature, slope length, lithology, distance from the road, distance from the stream, topographic wetness index (TWI), and sediment transport index (STI). The findings of this study have shown that both GAN and SMOTE data balancing approaches have helped to improve the accuracy of machine learning models. According to AUROC, the GAN method was able to boost the models by reaching the maximum accuracy of ANN (0.918), RF (0.933), DT (0.927), kNN (0.878), and SVM (0.907) when default parameters used. With the optimum parameters, all models performed best with GAN at their highest accuracy of ANN (0.927), RF (0.943), DT (0.923) and kNN (0.889), except SVM obtained the highest accuracy of (0.906) with SMOTE. Our finding suggests that RF balanced with GAN can provide the most reasonable criterion for landslide prediction. This research indicates that landslide data balancing may substantially affect the predictive capabilities of machine learning models. Therefore, the issue of DI in the spatial prediction of landslides should not be ignored. Future studies could explore other generative models for landslide data balancing. By using state-of-the-art GAN, the proposed model can be considered in the areas where the data are limited or imbalanced.

Landslide Susceptibility Analysis using Gradient Boosting Models: A Case Study in Penang Island, Malaysia

Tree-based gradient boosting (TGB) models gain popularity in various areas due to their powerful prediction ability and fast processing speed. This study aims to compare the landslide spatial prediction performance of TGB models and non-tree-based machine learning (NML) models in Penang Island, Malaysia. Two specific instances of TGB models, eXtreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) and two specific instances of NML models, artificial neural network (ANN) and support vector machine (SVM), are applied to make predictions of landslide susceptibility. Feature selection and oversampling techniques are considered to improve the prediction performance as well. The results are analyzed and discussed mainly based on receiver operating characteristic (ROC) curves as well as the area under the curves (AUC). The results show that TGB models give better prediction performance compared to NML models, no matter what the sample size is. The TGB models’ performances are improved when training with the dataset considering either feature selection or oversampling techniques. The highest AUC value of 0.9525 is obtained from the combination of XGBoost and SMOTE. The landslide susceptibility maps (LSMs) produced by XGBoost and LightGBM can provide valuable information in landslide management and mitigation in Penang Island, Malaysia.

Spatial prediction of landslide susceptibility using projected storm rainfall and land use in Himalayan region

Landslide susceptibility assessment (LSA) is a method used to reduce landslide vulnerability defined as landslide spatial prediction with the help of associative factors. The goal of the analysis is to forecast potential (2040, 2060, 2080, and 2100 AC) rainfall and land use and land cover (LULC) with the aid of the CSIRO-MK3.6.0 General circulation models (GCMs) climatic model and the dynamic conversion of land use and its effects (Dyna-CLUE) model. The purpose of this work is to produce landslide susceptibility map (LSM) of potential cycles correlated with expected rainfall and LULC data utilizing the binary logistic regression (BLR) models in the Upper Rangit River basin of eastern Himalayan region, India. Including rainfall and LULC total nineteen factors have been incorporated, these are mainly topographical, hydrological, geological, and environmental factors. A total 671 landslide locations have been mapped which divided randomly as training (70%) and validation (30%) datasets of LSM. Current (2018) LSM was validated by 30% of landslide inventory location and the result of area under curve of receiver operating characteristic (ROC) indicated that the LSM has 92.4% and 89.6% of the success and prediction rate respectively. To demonstrate the accuracy of the BLR landslide susceptibility (LS) model, this model was used to generated future LSMs based on projected of rainfall and LULC. The result of projected representative concentration pathway (RCP)-based rainfall depicted that the increasing trend of rainfall in the future period and the moderate to very high LS zones has also increased from 2040 to 2100. This study will be used to further study of landslide hazard (LH) studies with climatic approaches, and will also contribute to regional planning and development of current and future Upper Rangit River basin.

Spatial prediction of landslides along National Highway-6, Hoa Binh province, Vietnam using novel hybrid models

Landslides are considered to be a significant risk to life and property all over the world in general and in Vietnam in particular. Spatial prediction of landslides is required to reduce the landslides risk and to plan the development of hilly areas. In this regard, the accurate landslide susceptibility maps are very useful tool for decision-makers to identify areas where new landslides are likely to occur for planning timely adequate remedial measures. For the development of landslide susceptibility maps, seven hybrid models were developed namely AdaBoost-LMT (ABLMT), bagging-LMT (BLMT), cascade generalization-LMT (CGLMT), dagging-LMT (DLMT), MultiBoostAB-LMT (MBLMT), rotation forest-LMT (RFLMT) and random sub-space-LMT (RSSLMT) with logistic model trees (LMTs) as a base classifier. The model’s performance and validation were assessed through various statistical indices, such as sensitivity (SST), specificity (SPF), accuracy (ACC), area under ROC curve (AUC), RMSE and k index. The results show that all these models are performing well for the prediction of landslide susceptibility in the study area, but the performance of the RSSLMT model is the best (AUC: 0.816). In this study, open-source data has been used for the development of landslide susceptibility maps Along National Highway-6, passing through Hoa Binh province, Vietnam. These approaches can be applied also in other hilly regions of the world which are susceptible to landslides for better landslides prevention and management.

Three oversampling methods applied in a comparative landslide spatial research in Penang Island, Malaysia

Two main problems in landslide spatial prediction research are the lack of landslide samples (minority) to train the models and the misunderstanding of assigning equal costs to different misclassifications. In order to handle the problems properly, the research is conducted based on two main objectives, which are to augment the landslide sample data in an efficient way and to assign proper unequal costs to the two types of error when training and evaluating models. Resampling techniques, including random oversampling technique, synthetic minority oversampling technique and self-creating oversampling technique (SCOTE), are used to augment the minority class samples. Logistic regression (LR) and support vector machine (SVM) are used for landslide spatial classification. Receiver operating characteristic and cost curves are used to evaluate the models. The results show that the SVM models trained using the dataset generated by SCOTE with sample size of 10,000 have the best prediction performance. The nonparametric test, Kruskal–Wallis test, is used to test the difference of sample size between different groups, which shows that LR models are more sensitive to the change of sample size. Two landslide susceptibility maps are produced based on the models with the best prediction performance. The verification results show that the maps both successfully predict more than 86% of the susceptible area, which can provide valid information on landslide mitigation and prediction to the local authorities.

Potential of Ensemble Learning to Improve Tree-Based Classifiers for Landslide Susceptibility Mapping

Ensemble learning methods have been widely used due to their remarkable generalized performance, but their potential in landslide spatial prediction application is not fully studied. To take full advantage of ensemble learning techniques, the classification and regression tree classifier and four tree-based ensemble classifiers of random forest, extremely randomized tree, gradient boosting decision trees, and extreme gradient boosting decision trees are used in this study for landslide susceptibility assessment. Specifically, a stacking ensemble learning framework coupled with embedded feature selection is presented, consisting of multiple tree-based classifiers mentioned previously as base learners and logistic regression as a metalearner in a two-layer structure. In the study area of Yongxin, China, 364 historical landslide locations were first randomly partitioned into a ratio of 7/3 for training and testing the model. Then, a spatial database of 16 landslide causative factors was constructed for landslide prediction. Meanwhile, the relative importance of these factors were quantified by using the total number of feature splits and the average Gini index during the training process, and a novel embedded feature selection method was used in the base learner of the proposed framework to further improve the computational efficiency and predictive performance by allowing each base learner to obtain its own optimal subfeature space. Finally, different methods were assessed by using several evaluation criteria. Experimental results demonstrated that the proposed ensemble learning framework had the highest area under the curve value of 0.864, and it is more effective than the conventional tree-based classifiers and other ensemble learning methods.

Spatial Prediction of Landslides Using Hybrid Integration of Artificial Intelligence Algorithms with Frequency Ratio and Index of Entropy in Nanzheng County, China

The main object of this study is to introduce hybrid integration approaches that consist of state-of-the-art artificial intelligence algorithms (SysFor) and two bivariate models, namely the frequency ratio (FR) and index of entropy (IoE), to carry out landslide spatial prediction research. Hybrid integration approaches of these two bivariate models and logistic regression (LR) were used as benchmark models. Nanzheng County was considered as the study area. First, a landslide distribution map was produced using news reports, interpreting satellite images and a regional survey. A total of 202 landslides were identified and marked. According to the previous studies and local geological environment conditions, 16 landslide conditioning factors were chosen for landslide spatial prediction research: elevation, profile curvature, plan curvature, slope angle, slope aspect, stream power index (SPI), topographic wetness index (TWI), sediment transport index (STI), distance to roads, distance to rivers, distance to faults, lithology, rainfall, soil, normalized different vegetation index (NDVI), and land use. Then, the 202 landslides were randomly segmented into two parts with a ratio of 70:30. Seventy percent of the landslides (141) were used as the training dataset and the remaining landslides (61) were used as the validating dataset. Next, the evaluation models were built using the training dataset and compared by the receiver operating characteristics (ROC) curve. The results showed that all models performed well; the FR_SysFor model exhibited the best prediction ability (0.831), followed by the IoE_SysFor model (0.819), IoE_LR model (0.702), FR_LR model (0.696), IoE model (0.691), and FR model (0.681). Overall, these six models are practical tools for landslide spatial prediction research and the results can provide a reference for landslide prevention and control in the study area.

Spatial prediction strategy for landslides triggered by large earthquakes oriented to emergency response, mid-term resettlement and later reconstruction

To enhance the timeliness and accuracy of spatial prediction of co-seismic landslides, this paper proposes a prediction strategy in three phases: emergency response, mid-term resettlement and later reconstruction in the affected area. Taking the Lushan Ms 7.0 earthquake of 2013 as an example, it is stated as follows: In the emergency response phase, combining with topographic and geological data of the study area, the rapid assessment is conducted by the Newmark method without co-seismic landslide data. In the temporary resettlement phase with partial landslide data available, considering 10 factors including elevation, slope gradient, slope direction, topographic wetness index (TWI), peak ground acceleration (PGA), distances to the epicenter, distances to rivers, distances to roads, lithology and annual precipitation, a logistic regression (LR) model is established, and the spatial prediction of landslides is made. In the reconstruction phase with a complete landslide inventory, the incomplete data is replaced and the latest assessment is conducted based on the LR model. The accuracies of three phases are tested by using complete landslide data. The results show that the area under curve (AUC) values of the three stages is 0.59, 0.84 and 0.87, respectively. Such results can meet the requirement of spatial prediction of seismic landslides in each stage of disaster reduction.