Effects of Landslide Sampling Strategies on the Prediction Skill of Landslide Susceptibility Modelings

In this study, landslide susceptibility assessments were achieved using logistic regression, in a 523 km² area around the Eastern Mediterranean region of Southern Turkey. In reliable landslide susceptibility modeling, among others, an appropriate landslide sampling technique is always essential. In susceptibility assessments, two different random selection methods, ranging 78–83% for the train and 17–22% validation set in landslide affected areas, were applied. For the first, the landslides were selected based on their identity numbers considering the whole polygon while in the second, random grid cells of equal size of the former one was selected in any part of the landslides. Three random selections for the landslide free grid cells of equal proportion were also applied for each of the landslide affected data set. Among the landslide preparatory factors; geology, landform classification, land use, elevation, slope, plan curvature, profile curvature, slope length factor, solar radiation, stream power index, slope second derivate, topographic wetness index, heat load index, mean slope, slope position, roughness, dissection, surface relief ratio, linear aspect, slope/aspect ratio have been considered. The results showed that the susceptibility maps produced using the random selections considering the entire landslide polygons have higher performances by means of success and prediction rates.     

Landslide susceptibility mapping of a catchment area using frequency ratio,fuzzy logic and multivariate logistic regression approaches

Geospatial database creation for landslide susceptibility mapping is often an almost inhibitive activity. This has been the reason that for quite some time landslide susceptibility analysis was modelled on the basis of spatially related factors. This paper presents the use of frequency ratio, fuzzy logic and multivariate regression models for landslide susceptibility mapping on Cameron catchment area, Malaysia, using a Geographic Information System (GIS) and remote sensing data. Landslide locations were identified in the study area from the interpretation of aerial photographs, high resolution satellite images, inventory reports and field surveys. Topographical, geological data and satellite images were collected, processed, and constructed into a spatial database using GIS and image processing tools. There were nine factors considered for landslide susceptibility mapping and the frequency ratio coefficient for each factor was computed. The factors chosen that influence landslide occurrence were: topographic slope, topographic aspect, topographic curvature and distance from drainage, all from the topographic database; lithology and distance from lineament, taken from the geologic database; land cover from TM satellite image; the vegetation index value from Landsat satellite images; and precipitation distribution from meteorological data. Using these factors the fuzzy membership values were calculated. Then fuzzy operators were applied to the fuzzy membership values for landslide susceptibility mapping. Further, multivariate logistic regression model was applied for the landslide susceptibility. Finally, the results of the analyses were verified using the landslide location data and compared with the frequency ratio, fuzzy logic and multivariate logistic regression models. The validation results showed that the frequency ratio model (accuracy is 89%) is better in prediction than fuzzy logic (accuracy is 84%) and logistic regression (accuracy is 85%) models. Results show that, among the fuzzy operators, in the case with “gamma” operator (λ = 0.9) showed the best accuracy (84%) while the case with “or” operator showed the worst accuracy (69%).     

Using InfoVal method and GIS techniques for the spatial modelling of landslide susceptibility in the upper catchment of river Meenachil in Kerala

A GIS-based statistical methodology for landslide susceptibility zonation is described and its application to a study area in the Western Ghats of Kerala (India) is presented. The study area was approximately 218.44 km² and 129 landslides were identified in this area. The environmental attributes used for the landslide susceptibility analysis include geomorphology, slope, aspect, slope length, plan curvature, profile curvature, elevation, drainage density, distance from drainages, lineament density, distance from lineaments and land use. The quantitative relationship between landslides and factors affecting landslides are established by the data driven-Information Value (InfoVal) — method. By applying and integrating the InfoVal weights using ArcGIS software, a continuous scale of numerical indices (susceptibility index) is obtained with which the study area is divided into five classes of landslide susceptibility. In order to validate the results of the susceptibility analysis, a success rate curve was prepared. The map obtained shows that a great majority of the landslides (74.42%) identified in the field were located in susceptible and highly susceptible zones (27.29%). The area ratio calculated by the area under curve (AUC) method shows a prediction accuracy of 80.45%. The area having a high scale of susceptibility lies on side slope plateaus and denudational hills with high slopes where drainage density is relatively low and terrain modification is relatively intense.     

Spatial prediction of landslide susceptibility by combining evidential belief function,logistic regression and logistic model tree

Abstract

In this study, we introduced novel hybrid of evidence believe function (EBF) with logistic regression (EBF-LR) and logistic model tree (EBF-LMT) for landslide susceptibility modelling. Fourteen conditioning factors were selected, including slope aspect, elevation, slope angle, profile curvature, plan curvature, topographic wetness index (TWI), stream sediment transport index (STI), stream power index (SPI), distance to rivers, distance to faults, distance to roads, lithology, normalized difference vegetation index (NDVI), and land use. The importance of factors was assessed using correlation attribute evaluation method. Finally, the performance of three models was evaluated using the area under the curve (AUC). The validation process indicated that the EBF-LMT model acquired the highest AUC for the training (84.7%) and validation (76.5%) datasets, followed by EBF-LR and EBF models. Our result also confirmed that combination of a decision tree-logistic regression-based algorithm with a bivariate statistical model lead to enhance the prediction power of individual landslide models.     

Optimizing landslide susceptibility mapping in the Kongtong District,NW China: comparing the subdivision criteria of factors

Abstract

The standards applied to reclassify landslide-conditioning factors differ among studies and may change the accuracy of identifying landslide-prone areas. Therefore, we identified two standards per factor (elevation, aspect, slope, proximity to roads and proximity to streams) from the existing literature and set them as predisposing criteria in this paper. In addition to the five factors, lithology represented by types and a landslide inventory map produced from field surveys were also used in mapping. Thirty-two landslide susceptibility maps were generated based on weights-of-evidence and evaluated using the relative operative characteristic method. The results show that the subdivision criteria of factors change the accuracy, with the success rate varying from 84.34% to 87.51%. The map with the highest value captures more landslides in relatively higher susceptibility classes and is therefore considered the optimal one. Ultimately, a simplified mode of combining subdivision criteria is proposed to simplify comparison.     

Detecting patterns of climate change in long-term forecasts of marine environmental parameters

ABSTRACT

Forecasting environmental parameters in the distant future requires complex modelling and large computational resources. Due to the sensitivity and complexity of forecast models, long-term parameter forecasts (e.g. up to 2100) are uncommon and only produced by a few organisations, in heterogeneous formats and based on different assumptions of greenhouse gases emissions. However, data mining techniques can be used to coerce the data to a uniform time and spatial representation, which facilitates their use in many applications. In this paper, streams of big data coming from AquaMaps and NASA collections of 126 long-term forecasts of nine types of environmental parameters are processed through a cloud computing platform in order to (i) standardise and harmonise the data representations, (ii) produce intermediate scenarios and new informative parameters, and (iii) align all sets on a common time and spatial resolution. Time series cross-correlation applied to these aligned datasets reveals patterns of climate change and similarities between parameter trends in 10 marine areas. Our results highlight that (i) the Mediterranean Sea may have a standalone ‘response’ to climate change with respect to other areas, (ii) the Poles are most representative of global forecasted change, and (iii) the trends are generally alarming for most oceans.     

A comparison of Support Vector Machines and Bayesian algorithms for landslide susceptibility modelling

Abstract

In this study, the main goal is to compare the predictive capability of Support Vector Machines (SVM) with four Bayesian algorithms namely Naïve Bayes Tree (NBT), Bayes network (BN), Naïve Bayes (NB), Decision Table Naïve Bayes (DTNB) for identifying landslide susceptibility zones in Pauri Garhwal district (India). First, landslide inventory map was built using 1295 historical landslide data, then in total sixteen influencing factors were selected and tested for landslide susceptibility modelling. Performance of the model was evaluated and compared using Statistical based index methods, Area under the Receiver Operating Characteristic (ROC) curve named AUC, and Chi-square method. Analysis results show that that the SVM has the highest prediction capability, followed by the NBT, DTNBT, BN and NB, respectively. Thus, this study confirms that the SVM is one of the benchmark models for the assessment of susceptibility of landslides.     

Landslide susceptibility mapping by remote sensing and geomorphological data: case studies on the Sorrentina Peninsula (Southern Italy)

The Sorrentina Peninsula is a densely populated area with high touristic impact. It is located in a morphologically complex zone of Southern Italy frequently affected by dangerous and calamitous landslides. This work contributes to the prevention of such natural disasters by applying a GIS-based interdisciplinary approach aimed to map the areas more potentially prone to trigger slope instability phenomena. We have developed the Landslide Susceptibility Index (LSI) combining five weighted and ranked susceptibility parameters on a GIS platform. These parameters are recognized in the literature as the main predisposing factors for triggering landslides. This work combines analyses conducted on Remote Sensing, Geo-Lithology and Morphometry data and it is organized in the following logical steps: i) Multi-temporal InSAR technique was applied to Envisat-ASAR (2003–2010) and COSMO-SkyMed (2013–2015) datasets to obtain the ground displacement time series and the relative mean ground velocity maps. InSAR allowed the detection of the areas that are subjected to ground deformation and the main affected municipalities; ii) Such deformation areas were investigated through airborne photo interpretation to identify the presence of geomorphological peculiarities connected to potential slope instability. Subsequently, some of these peculiarities were checked on the field; iii) In these deformation areas the susceptibility parameters were mapped in the entire territory of Amalfi and Conca dei Marini and then investigated with a multivariate analysis to derive the classes and the respective weights used in the LSI calculation. The resulting LSI map classifies the two municipalities with high spatial resolution (2m) according to five classes of instability. The map highlights that the high/very high susceptibility zones cover 6% of the investigated territory and correspond to potential landslide source areas characterized by 25°-70° slope angles. A spatial analysis between the map of the historical landslides and the areas classified according to susceptibility allowed testing of the reliability of the LSI Index, resulting in 85% prediction accuracy.     

Multi-hazard assessment using machine learning and remote sensing in the North Central region of Vietnam

Natural hazards constitute a diverse category and are unevenly distributed in time and space. This hinders predictive efforts, leading to significant impacts on human life and economies. Multi-hazard prediction is vital for any natural hazard risk management plan. The main objective of this study was the development of a multi-hazard susceptibility mapping framework, by combining two natural hazards—flooding and landslides—in the North Central region of Vietnam. This was accomplished using support vector machines, random forest, and AdaBoost. The input data consisted of 4591 flood points, 1315 landslide points, and 13 conditioning factors, split into training (70%), and testing (30%) datasets. The accuracy of the models' predictions was evaluated using the statistical indices root mean square error, area under curve (AUC), mean absolute error, and coefficient of determination. All proposed models were good at predicting multi-hazard susceptibility, with AUC values over 0.95. Among them, the AUC value for the support vector machine model was 0.98 and 0.99 for landslide and flood, respectively. For the random forest model, these values were 0.98 and 0.98, and for AdaBoost, they were 0.99 and 0.99. The multi-hazard maps were built by combining the landslide and flood susceptibility maps. The results showed that approximately 60% of the study area was affected by landslides, 30% by flood, and 8% by both hazards. These results illustrate how North Central is one of the regions of Vietnam that is most severely affected by natural hazards, particularly flooding, and landslides. The proposed models adapt to evaluate multi-hazard susceptibility at different scales, although expert intervention is also required, to optimize the algorithms. Multi-hazard maps can provide a valuable point of reference for decision makers in sustainable land-use planning and infrastructure development in regions faced with multiple hazards, and to prevent and reduce more effectively the frequency of floods and landslides and their damage to human life and property.     

Comparison of landslide susceptibility mapping based on statistical index,certainty factors,weights of evidence and evidential belief function models

The main aim of this study was to produce landslide susceptibility maps using statistical index (SI), certainty factors (CF), weights of evidence (WoE) and evidential belief function (EBF) models for the Long County, China. Firstly, a landslide inventory map, including a total of 171 landslides, was compiled on the basis of earlier reports, interpretation of aerial photographs and supported by extensive field surveys. Thereafter, all landslides were randomly separated into two data sets: 70% landslides (120 points) were selected for establishing the model and the remaining landslides (51 points) were used for validation purposes. Eleven landslide conditioning factors, such as slope aspect, slope angle, plan curvature, profile curvature, altitude, distance to faults, distance to roads, distance to rivers, lithology, NDVI and land use, were considered for landslide susceptibility mapping in this study. Then, the SI, CF, WoE and EBF models were used to produce the landslide susceptibility maps for the study area. Finally, the four models were validated using area under the curve (AUC) method. According to the validation results, the EBF model (AUC = 78.93%) has a higher prediction accuracy than the SI model (AUC = 77.72%), the WoE model (AUC = 77.62%) and the CF model (AUC = 77.72%). Similarly, the validation results also indicate that the EBF model has the highest training accuracy of 80.25%, followed by SI (79.80%), WoE (79.71%) and CF (79.67%) models.     

Remote Sensing and GIS Based Landslide Susceptibility Assessment using Binary Logistic Regression Model: A Case Study in the Ganeshganga Watershed, Himalayas

A comprehensive Landslide Susceptibility Zonation (LSZ) map is sought for adopting any landslide preventive and mitigation measures. In the present study, LSZ map of landslide prone Ganeshganga watershed (known for Patalganga Landslide) has been generated using a binary logistic regression (BLR) model. Relevant thematic layers pertaining to the causative factors for landslide occurrences, such as slope, aspect, relative relief, lithology, tectonic structures, lineaments, land use and land cover, distance to drainage, drainage density and anthropogenic factors like distance to road, have been generated using remote sensing images, field survey, ancillary data and GIS techniques. The coefficients of the causative factors retained by the BLR model along with the constant have been used to construct the landslide susceptibility map of the study area, which has further been categorized into four landslide susceptibility zones from high to very low. The resultant landslide susceptibility map was validated using receiver operator characteristic (ROC) curve analysis showing an accuracy of 95.2 % for an independent set of test samples. The result also showed a strong agreement between distribution of existing landslides and predicted landslide susceptibility zones.     

Geographically weighted elastic net logistic regression

This paper develops a localized approach to elastic net logistic regression, extending previous research describing a localized elastic net as an extension to a localized ridge regression or a localized lasso. All such models have the objective to capture data relationships that vary across space. Geographically weighted elastic net logistic regression is first evaluated through a simulation experiment and shown to provide a robust approach for local model selection and alleviating local collinearity, before application to two case studies: county-level voting patterns in the 2016 USA presidential election, examining the spatial structure of socio-economic factors associated with voting for Trump, and a species presence–absence data set linked to explanatory environmental and climatic factors at gridded locations covering mainland USA. The approach is compared with other logistic regressions. It improves prediction for the election case study only which exhibits much greater spatial heterogeneity in the binary response than the species case study. Model comparisons show that standard geographically weighted logistic regression over-estimated relationship non-stationarity because it fails to adequately deal with collinearity and model selection. Results are discussed in the context of predictor variable collinearity and selection and the heterogeneities that were observed. Ongoing work is investigating locally derived elastic net parameters.     

Application of frequency ratio,weights of evidence and evidential belief function models in landslide susceptibility mapping

The landslide hazard occurred in Taibai County has the characteristics of the typical landslides in mountain hinterland. The slopes mainly consist of residual sediments and locate along the highway. Most of them are in the less stable state and in high risk during rainfall in flood season especially. The main purpose of this paper is to produce landslide susceptibility maps for Taibai County (China). In the first stage, a landslide inventory map and the input layers of the landslide conditioning factors were prepared in the geographic information system supported by field investigations and remote sensing data. The landslides conditioning factors considered for the study area were slope angle, altitude, slope aspect, plan curvature, profile curvature, distance to faults, distance to rivers, distance to roads, normalized difference vegetation index, lithological unit, rainfall and land use. Subsequently, the thematic data layers of conditioning factors were integrated by frequency ratio (FR), weights of evidence (WOE) and evidential belief function (EBF) models. As a result, landslide susceptibility maps were obtained. In order to compare the predictive ability of these three models, a validation procedure was conducted. The curves of cumulative area percentage of ordered index values vs. the cumulative percentage of landslide numbers were plotted and the values of area under the curve (AUC) were calculated. The predictive ability was characterized by the AUC values and it indicates that all these models considered have relatively similar and high accuracies. The success rate of FR, WOE and EBF models was 0.9161, 0.9132 and 0.9129, while the prediction rate of the three models was 0.9061, 0.9052 and 0.9007, respectively. Considering the accuracy and simplicity comprehensively, the FR model is the optimum method. These landslide susceptibility maps can be used for preliminary land use planning and hazard mitigation purpose.     

粒子群优化BP神经网络的滑坡敏感性评价

滑坡敏感性评价是地质灾害预测预报的关键环节。针对BP神经网络易陷入局部最小值、收敛速度慢等问题,该文以三峡库区秭归县境内为研究区,采用粒子群优化(PSO)算法对BP神经网络的初始权值和阈值进行优化,构建PSO-BP神经网络滑坡敏感性预测模型,实现研究区滑坡敏感性评价。采用受试者工作特征曲线分析模型预测精度,得到PSO-BP神经网络预测精度为0.931,预测结果与实际滑坡总体空间分布具有良好的一致性,且预测能力优于BP神经网络。实验结果表明,PSO-BP神经网络耦合模型在实现滑坡敏感性评价上具有理想的预测精度和良好的适用性。     

高光谱遥感影像多级联森林深度网络分类算法

高光谱遥感技术在环境监测、应急保障、精细地物提取等方面有着广泛的应用,随着高分五号高光谱数据的正式发布,高光谱遥感技术将发挥更重要的作用。遥感影像分类作为高光谱遥感影像信息处理的重要部分,已成为当前研究重点。本文针对传统多级联森林深度学习中模型复杂、无法利用基分类器差异信息、对类间差异较小的样本无法正确区分等不足,提出了一种改进的多级联森林深度学习模型,在模型框架中,分别采用了随机森林和旋转森林作为基分类器,并引入逻辑回归分类器作为判别器用于训练层扩展。相较于传统的深度神经网络,改进的多级联森林深度网络超参数较少且能够自适应确定训练层,更方便进行模型优化。实验采用了高分五号数据集及两个公开的高光谱数据集(Indian Pines数据集及Pavia University数据集)进行精度评定,同时选择了传统分类器支持向量机、深度置信网等模型作为对比分析。实验结果表明,改进的多级联森林深度学习模型能有效地进行高光谱遥感影像分类,且较传统的分类方法精度有所提升。     

CF耦合LR模型的地质灾害易发性评价

为对自然灾害频发的西北地区进行地质灾害易发性研究,本文考虑了高程、坡向等10类因素,利用确定性系数(CF)模型和确定性系数耦合逻辑回归(CF-LR)模型对城固县地质灾害易发性进行评价。结果表明,CF模型和CF-LR模型的灾害比均呈递增状态,高、极高灾害点密度分别为4.75、5.97个/km²,且两种模型的测试集受试者工作特征(ROC)曲线下面积(AUC)分别为0.812、0.835,验证集ROC和AUC值分别为0.862、0.891。两种模型均能有效评价区域内地质灾害易发性,且CF-LR模型具有更高的评价精度。     

顾及样本敏感性的滑坡易发性评价

滑坡作为一种危害极大的自然地质现象,严重威胁着人民的生命财产安全。因此,科学、准确地评价滑坡体的易发性至关重要。随着机器学习的发展,基于机器学习的滑坡易发性评价逐渐成为研究热点。而在真实情况中,滑坡区域与非滑坡区域面积占比悬殊,这使得机器学习模型的应用存在较严重的样本不均衡问题。本文采用样本敏感性分析方法,综合多个机器学习模型在不同比例的正负滑坡样本集上的表现,以获取最均衡滑坡样本集;并在此样本集基础上采用深度随机森林模型,在示范研究区开展滑坡易发性评价。最终的评价结果接近真实分布,表明本文方法具有较好的有效性。     

基于GIS的Logistic回归模型在区域滑坡敏感性评价中的应用

本文基于GIS技术和Logistic回归模型进行滑坡敏感性评价定量分析方法,并以江苏省连云港市郊区为研究区域,建立了地质、地形数据库等滑坡因子空间数据库和滑坡空间分布数据库,进行了滑坡影响因子敏感性分析。对连云港市郊区滑坡灾害在空间上的预测结果具有重要的现实意义,对推广应用、防灾减灾具有实际的指导意义。     

How many samples are needed? An investigation of binary logistic regression for selective omission in a road network

ABSTRACT

Selective omission in a road network (or road selection) means to retain more important roads, and it is a necessary operator to transform a road network at a large scale to that at a smaller scale. This study discusses the use of the supervised learning approach to road selection, and investigates how many samples are needed for a good performance of road selection. More precisely, the binary logistic regression is employed and three road network data with different sizes and different target scales are involved for testing. The different percentages and numbers of strokes are randomly chosen for training a logistic regression model, which is further applied into the untrained strokes for validation. The performances of using the different sample sizes are mainly evaluated by an error rate estimate. Significance tests are also employed to investigate whether the use of different sample sizes shows statistically significant differences. The experimental results show that in most cases, the error rate estimate is around 0.1–0.2; more importantly, only a small number (e.g., 50–100) of training samples is needed, which indicates the usability of binary logistic regression for road selection.     

Landslide Susceptibility Mapping Using Remotely Sensed Data through Conditional Probability Analysis Using Seed Cell and Point Sampling Techniques

Rapid urbanization, intense infra-structure development and increased tourism related activities have resulted in the change of landscape of the Kodaikkanal town and its surrounding, a popular hill town in Tamilnadu, South India. As an after effect, the numbers of landslides and rock-falls have increased steadily in the past decade. Landslide susceptibility analysis is carried out for this area using conditional probability analysis. The geo-spatial database for mapping landslide susceptibility consists of the factors - Relief, Slope, Aspect, Curvature, Weathering, Land use, Topographic Wetness Index and Proximity to road. Two sampling strategies ?C point and seed-cell are compared for landslide susceptibility mapping. The Landslide Susceptibility map developed using conditional probability method is verified using R index for both sampling strategies. The study shows that both the sampling strategies perform with good accuracy, seed cell technique excels slightly over point sampling. 86.11% of the landslides fall in the high and critical susceptible zones. The results show that conditional probability technique provides a simple tool for susceptibility analysis. The method can be used at regional scale and is a valuable input for planning purpose.