Application of binary logistic regression analysis and its validation for landslide susceptibility mapping in part of Garhwal Himalaya,India

Landslides cause heavy damage to property and infrastructure, in addition to being responsible for the loss of human lives, in many parts of the Himalaya. It is possible to take appropriate management measures to reduce the risk from potential landslide hazard with the help of landslide hazard zonation (LHZ) maps. The present work is an attempt to utilize binary logistic regression analysis for the preparation of a landslide susceptibility map for a part of Garhwal Himalaya, India, which is highly prone to landslides, by taking the geological, geomorphological and topographical parameters into consideration. Remote sensing and the geographic information system (GIS) were found to be very useful in the input database preparation, data integration and analysis stages. The coefficients of the predictor variables are estimated using binary logistic regression analysis and are used to calculate the landslide susceptibility for the entire study area within a GIS environment. The receiver operator characteristic curve analysis gives 88.7% accuracy for the developed model.     

基于多特征面向对象区域滑坡现象识别

高空间分辨率卫星遥感数据的发展为滑坡灾害数据获取和更新提供了新的途径。以西北黄土高原为研究区,提出了一种基于多特征面向对象区域滑坡现象的识别方法,基于单期高空间分辨率遥感数据,利用集合和特征组合进行区域滑坡现象识别,实验结果表明:该方法是识别滑坡现象有效的技术方法之一,对开展滑坡监测、影像理解和地学分析具有重要的研究意义。     

Using likelihood ratio functions for modeling the conditional probability of occurrence of future landslides for risk assessment

The most crucial and difficult task in landslide hazard analysis is estimating the conditional probability of occurrence of future landslides in a study area within a specific time period, given specific geomorphic and topographic features. This task can be addressed with a mathematical model that estimates the required conditional probability in two stages: “relative hazard mapping” and “empirical probability estimation.” The first stage divides the study area into a number of “prediction” classes according to their relative likelihood of occurrence of future landslides, based on the geomorphic and topographic data. Each prediction class represents a relative level of hazard with respect to other prediction classes. The number of classes depends on the quantity and quality of input data. Several quantitative models have been developed and tested for use in this stage; the objective is to delineate typical settings in which future landslides are likely to occur. In this stage, problems related to different degrees of resolution in the input data layers are resolved. The second stage is to empirically estimate the conditional probability of landslide occurrence in each prediction class by a cross-validation technique. The basic strategy is to divide past occurrences of landslides into two groups, a “modeling group” and a “validation group”. The first mapping stage is repeated, but the prediction is limited to only those landslide occurrences in the modeling group that are used to construct a new set of prediction classes. The new set of prediction classes is compared to the distribution of landslide occurrences in the validation group. Statistics from the comparison provide a quantitative measure of the conditional probability of occurrence of future landslides.     

Probabilistic landslide hazard mapping using GIS and remote sensing data at Boun,Korea

The aim of this study is to evaluate the hazard of landslides at Boun, Korea, using a Geographic Information System (GIS) and remote sensing. Landslide locations were identified in the Boun area from interpretation of aerial photographs and field surveys. The topographic, soil, forest, geologic, lineament and land cover data were collected, processed and constructed into a spatial database using GIS and remote sensing data. The factors that influence landslide occurrence, such as slope, aspect and curvature of the topography, were calculated from the topographic database. Texture, material, drainage and effective soil thickness were extracted from the soil database, and type, age, diameter and density of timber were extracted from the forest database. The lithology was extracted from the geological database and lineaments were detected from Indian Remote Sensing (IRS) satellite images. The land cover was classified based on the Landsat Thematic Mapper (TM) satellite image. Landslide hazard areas were analysed and mapped, using the landslide-occurrence factors, by the probability–likelihood ratio method. The results of the analysis were verified using actual landslide location data. The validation results showed satisfactory agreement between the hazard map and the existing data on landslide locations.     

A distributed framework for multi-risk assessment of natural hazards used to model the effects of forest fire on hydrology and sediment yield

Within the European Commission-funded MEDIGRID project, Grid computing technology is used to integrate various natural hazard models and data sets, maintained independently at different centres in Europe, into a single system, accessible to users over the internet. Each centre forms a process (application) or data storage node and has been fitted with the Globus toolkit, which provides the distributed computing environment functionality that is required for the system set up. In addition, several Grid data management components were developed to allow the system to operate on different computing platforms. Access to the data and application management services is enabled through a Grid Portal. A series of portlets enable users to access the system, providing a personalised interface to the Grid. Integration of the individual models required them to be modified as web services, so as to be run remotely over the internet. As the models have different data characteristics, a common data format was adopted for creating harmonised data sets and allowing the exchange of data between the models. As an example, the Fire Spread Engine model is used to derive a map of areas that have been burnt by fire. This forms an input to the SHETRAN hydrology, soil erosion and landslide model, which in turn could provide data for other models such as vegetation regeneration. The use of the system is demonstrated for a site in south-west Spain where a large forest fire occurred on 2 August 2003. The MEDIGRID system marks an advance in the integration of independently constructed models to provide improved hazard assessment technology.     

A Object-oriented Landslide Extraction Method based on Morphological Opening Operation

Landslide is one of the most common geological disasters,and it is of great significance to quickly and accurately obtain the hazard degree and distribution of landslide.By introducing the morphological opening operation and the regional level set algorithm to construct the object\|oriented landslide extraction method,and using the multi\|spectral image of GF\|2 satellite as the data source,the landslide extraction experiment is carried out by using the constructed method and the landslide extraction method based on the pixel in the study area of the Bagmati area in Nepal,and the results of the landslide extraction are analyzed.The experimental results show that the object\|oriented landslide extraction method is more accurate than the extraction method based on the pixel,and the anti\|interference ability of the clouds and snows is stronger than that of the pixel\|based landslide extraction method.     

Landslide hazard and risk analyses at a landslide prone catchment area using statistical based geospatial model

This paper presents the application of remote sensing techniques, digital image analysis and Geographic Information System tools to delineate the degree of landslide hazard and risk areas in the Balik Pulau area in Penang Island, Malaysia. Its causes were analysed through various thematic attribute data layers for the study area. Firstly, landslide locations were identified in the study area from the interpretation of aerial photographs, satellite imageries, field surveys, reports and previous landslide inventories. Topographic, geologic, soil and satellite images were collected and processed using Geographic Information System and image processing tools. There are 12 landslide-inducing parameters considered for the landslide hazard analyses. These parameters are: topographic slope, topographic aspect, plan curvature, distance to drainage and distance to roads, all derived from the topographic database; geology and distance to faults, derived from the geological database; landuse/landcover, derived from Landsat satellite images; soil, derived from the soil database; precipitation amount, derived from the rainfall database; and the vegetation index value, derived from SPOT satellite images. In addition, hazard analyses were performed using landslide-occurrence factors with the aid of a statistically based frequency ratio model. Further, landslide risk analysis was carried out using hazard map and socio-economic factors using a geospatial model. This landslide risk map could be used to estimate the risk to population, property and existing infrastructure like transportation networks. Finally, to check the accuracy of the success-rate prediction, the hazard map was validated using the area under curve method. The prediction accuracy of the hazard map was 89%. Based on these results the authors conclude that frequency ratio models can be used to mitigate hazards related to landslides and can aid in land-use planning.     

Ensemble of extreme learning machine for landslide displacement prediction based on time series analysis

Landslide hazard is a complex nonlinear dynamical system with uncertainty. The evolution of landslide is influenced by many factors such as tectonic, rainfall and reservoir level fluctuation. Using a time series model, total accumulative displacement of landslide can be divided into the trend component displacement and the periodic component displacement according to the response relation between dynamic changes in landslide displacement and inducing factors. In this paper, a novel neural network technique called ensemble of extreme learning machine (E-ELM) is proposed to investigate the interactions of different inducing factors affecting the evolution of landslide. Grey relational analysis is used to sieve out the more influential inducing factors as the inputs in E-ELM. Trend component displacement and periodic component displacement are forecasted, respectively; then, total predictive displacement is obtained by adding the calculated predictive displacement value of each sub. Performances of our model are evaluated by using real data from Baishuihe landslide in the Three Gorges Reservoir of China, and it provides a good representation of the measured slide displacement behavior.     

Earthquake-induced landslide dam in the Kashmir Himalayas

The Kashmir earthquake of 2005 triggered numerous landslides in inaccessible areas of the western Himalayas, which could be mapped using satellite remote sensing. The largest recorded seismicity-induced landslide dammed a river, which resulted in the formation of a stream in the toe region and created two reservoirs that pose an enormous threat in the event of a landslide dam breach. Using Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data sets corresponding to the pre- and post-earthquake period and derived digital elevation models, landslide-induced lakes were monitored. The aerial extent, depth profile and volume of both the reservoirs were determined. This study has demonstrated the utility of ASTER data in providing valuable information that is critical for hazard mitigation in case of a landslide dam breach.     

遥感技术在我国滑坡研究中的应用综述

中国滑坡遥感研究与应用已有30多年的发展历史,作为滑坡调查的主要手段在大型工程建设中的滑坡灾害调查及危险性评价中发挥了重要作用。从4方面阐述了遥感技术在我国滑坡研究中的应用：① 区域滑坡灾害遥感调查;② 大型单体滑坡遥感调查;③ 滑坡灾害遥感监测;④ 遥感应用于滑坡风险评估。随着遥感技术理论的逐步完善和遥感图像空间分辨率、时间分辨率与波谱分辨率的不断提高,遥感技术已成为滑坡灾害调查、动态监测与预警、灾情实时调查与损失评估等工作中不可缺少的重要手段之一。     

Landslide susceptibility mapping using downscaled AMSR-E soil moisture: A case study from Cleveland Corral, California, US

As soil moisture increases, slope stability decreases. Remotely sensed soil moisture data can provide routine updates of slope conditions necessary for landslide predictions. For regional scale landslide investigations, only remote-sensing methods have the spatial and temporal resolution required to map hazard increases. Here, a dynamic physically-based slope stability model that requires soil moisture is applied using remote-sensing products from multiple Earth observing platforms. The resulting landslide susceptibility maps using the advanced microwave scanning radiometer (AMSR-E) surface soil moisture are compared to those created using variable infiltration capacity (VIC-3L) modeled soil moisture at Cleveland Corral landslide area in California, US. Despite snow cover influences on AMSR-E surface soil moisture estimates, a good relationship between the downscaled AMSR-E's surface soil moisture and the VIC-3L modeled soil moisture is evident. The AMSR-E soil moisture mean (0.17 cm³/cm³) and standard deviation (0.02 cm³/cm³) are very close to the mean (0.21 cm³/cm³) and standard deviation (0.09 cm³/cm³) estimated by VIC-3L model. Qualitative results show that the location and extent of landslide prone regions are quite similar. Under the maximum saturation scenario, 0.42% and 0.49% of the study area were highly susceptible using AMSR-E and VIC-3L model soil moisture, respectively.     

Spatial prediction of shallow landslide using Bat algorithm optimized machine learning approach: A case study in Lang Son Province,Vietnam

This study develops a machine learning method that hybridizes the Least Squares Support Vector Classification (LSSVC) and Bat Algorithm (BA), named as BA-LSSVC, for spatial prediction of shallow landslide. To construct and verify the hybrid method, a Geographic Information System (GIS) database for the study area of Lang Son province (Vietnam) has been employed. LSSVC is used to separate data samples in the GIS database into two categories of non-landslide (negative class) and landslide (positive class). The BA metaheuristic is employed to assist the LSSVC model selection process by fine-tuning its hyper-parameters: the regularization coefficient and the kernel function parameter. Experimental results point out that the hybrid BA-LSSVC can help to achieve a desired prediction with an accuracy rate of more than 90%. The performance of BA-LSSVC is also better than those of benchmark methods, including the Convolutional Neural Network, Relevance Vector Machine, Artificial Neural Network, and Logistic Regression. Hence, the newly developed model is a capable tool to assist local authority in landslide hazard mitigation and management.     

滑坡稳定性与土地利用关联规则挖掘及应用 ——以三峡库区秭归到巴东段为例

三峡库区滑坡灾害广泛发育,其稳定性受土地利用变化等人类工程活动的影响。采用数据挖掘技术研究库区土地利用变化对滑坡稳定性的影响及其规律,利用三个时相的遥感影像得到实验区滑坡面上两个时段间的土地利用变化监测图,用Apriori算法挖掘出滑坡稳定性与土地利用变化类型之间的强规则,用马尔可夫链模型预测滑坡面上土地利用的变化趋势,将预测结果用于对滑坡稳定性发展的分析评估。通过实验分析,所采用的方法可用于预测滑坡稳定性变化趋势,为滑坡灾害的监测预警提供决策支持。     

基于图像处理的滑坡后缘裂缝运动轨迹分析

滑坡后缘裂缝运动轨迹分析对于滑坡监测预警至关重要,针对滑坡后缘已有裂缝带,提出了一种基于最大类间算法（OTSU）、Canny算子与目标物体特征识别的滑坡后缘裂缝检测方法,通过对滑坡后缘裂缝图像进行预处理、形态学处理与边缘检测、边缘轨迹比较,获取滑坡后缘裂缝位移情况。经测试,该方法在上位机（基于QT平台开发）上能有效检测滑坡后缘裂缝,并判别理论模型下的运动轨迹,可为滑坡灾害预警提供支撑。     

Assessing the activity of a large landslide in southern Italy by ground-monitoring and SAR interferometric techniques

Landslides are recognized as one of the most damaging natural hazards in Italy. Campania region represents a complex geological setting, where mass movements of different types are widespread, and urban expansion can be increasingly seen by the presence of buildings and infrastructure in landslide-prone areas. In such a context, monitoring of unstable slopes represents a key activity in the process of landslide risk prevention and mitigation, in order to correctly establish a cause–effect correlation and to predict the possible reactivation phases that may result in high costs for the human society. This article focuses on the application of different methods of landslide analysis and monitoring, including those developed more recently and based on data acquired by satellites and processed by synthetic aperture radar (SAR) interferometric techniques. The study area is a small town, Calitri, known worldwide for the large landslide reactivated by the 23 November 1980 earthquake that destroyed a large sector of the historical centre. The site has been investigated by two ground-monitoring campaigns, the analysis of which allowed identification of the evolution of landslide activity over time. Furthermore, differential SAR interferometry (DInSAR), based upon two different approaches, allowed us to produce point-wise and wide area deformation maps after processing data sets of Earth Resource Satellite 1/2 (ERS-1/2) images, respectively acquired in 1992–2001 and 1992–1995. The results obtained from this analysis highlighted the potentiality of remote-sensing tools in landslide hazard assessment and led to development of a research project based on the installation of corner reflectors along unstable slopes and aimed at creating a field–Earth observation monitoring system.     

Incorporating the effects of topographic amplification and sliding areas in the modeling of earthquake-induced landslide hazards,using the cumulative displacement method

A procedure that considers topographic effects and runout behavior is proposed for analyzing seismic landslide hazards. The theoretical topographic amplification factors and corresponding amplified ground motion are calculated. By using the amplified motion, a cumulative displacement map is generated through the Newmark's displacement method. The high displacement areas are defined as the source areas of landslides. A runout simulation that identifies sliding routes and the final deposition areas of the sliding material from these source areas is performed. Finally, the complete set of landslide zones, including source, and sliding and deposition areas, is predicted.The landslide hazard maps of the Mt. Baishiya region, Nantou, Taiwan are evaluated, and the maps of actual landslides triggered by the September 21st, 1999, Chi-Chi earthquake are compared with the prediction. The results show that the proposed procedure, which combines topographic effects and runout simulation, can generate more accurate predictions for seismic landslide hazard analysis. However, this slight improvement over the procedure that only considers topographic effects is within the uncertainty levels of the input parameters. This slight improvement is obtained by a relatively complex and time-consuming procedure, and further research is required towards evaluating the viability of the proposed model when topographic effects cannot be neglected.     

Temporal Feature Integration for Music Genre Classification

Temporal feature integration is the process of combining all the feature vectors in a time window into a single feature vector in order to capture the relevant temporal information in the window. The mean and variance along the temporal dimension are often used for temporal feature integration, but they capture neither the temporal dynamics nor dependencies among the individual feature dimensions. Here, a multivariate autoregressive feature model is proposed to solve this problem for music genre classification. This model gives two different feature sets, the diagonal autoregressive (DAR) and multivariate autoregressive (MAR) features which are compared against the baseline mean-variance as well as two other temporal feature integration techniques. Reproducibility in performance ranking of temporal feature integration methods were demonstrated using two data sets with five and eleven music genres, and by using four different classification schemes. The methods were further compared to human performance. The proposed MAR features perform better than the other features at the cost of increased computational complexity.     

不确定GM-CFSFDP聚类算法在滑坡危险性预测中的应用

针对滑坡危险性预测中降雨等不确定诱发因素难以有效处理,CFSFDP算法需要人工尝试设置密度阈值以及对大规模数据集无法进行准确聚类等问题,为了提高滑坡危险性预测准确度,提出一种基于网格与类合并的不确定CFSFDP （简称不确定GM-CFSFDP）聚类算法.该算法首先引入不确定数据处理方法,设计了E-ML距离公式,有效刻画降雨不确定因素;其次通过网格划分的思想把大规模数据集划分到多个网格空间中,实现大规模数据有效编码;计算网格平均密度,建立网格密度阈值分布模型,动态获得网格密度阈值;最后利用层次聚类思想对关联性较高的类进行合并,构建不确定GM-CFSFDP算法模型,在延安宝塔区进行滑坡实例验证.实验结果表明不确定GM-CFSFDP聚类算法获得较高的预测精度,从而验证了该算法在滑坡危险性预测中的可行性和先进性.     

Remote sensing of soil salinity: potentials and constraints

Soil salinity caused by natural or human-induced processes is a major environmental hazard. The global extent of primary salt-affected soils is about 955 M ha, while secondary salinization affects some 77 M ha, with 58% of these in irrigated areas. Nearly 20% of all irrigated land is salt-affected, and this proportion tends to increase in spite of considerable efforts dedicated to land reclamation. This requires careful monitoring of the soil salinity status and variation to curb degradation trends, and secure sustainable land use and management. Multitemporal optical and microwave remote sensing can significantly contribute to detecting temporal changes of salt-related surface features. Airborne geophysics and ground-based electromagnetic induction meters, combined with ground data, have shown potential for mapping depth of salinity occurrence. This paper reviews various sensors (e.g. aerial photographs, satellite- and airborne multispectral sensors, microwave sensors, video imagery, airborne geophysics, hyperspectral sensors, and electromagnetic induction meters) and approaches used for remote identification and mapping of salt-affected areas. Constraints on the use of remote sensing data for mapping salt-affected areas are shown related to the spectral behaviour of salt types, spatial distribution of salts on the terrain surface, temporal changes on salinity, interference of vegetation, and spectral confusions with other terrain surfaces.As raw remote sensing data need substantial transformation for proper feature recognition and mapping, techniques such as spectral unmixing, maximum likelihood classification, fuzzy classification, band ratioing, principal components analysis, and correlation equations are discussed. Lastly, the paper presents modelling of temporal and spatial changes of salinity using combined approaches that incorporate different data fusion and data integration techniques.     

Research on Wenchuan Earthquake-induced LandslidesRapid Recognition from ZY-3 Imagery

Presently,regional earthquake-induced landslides is mainly obtained by field survey and visual interpretation from remote sensing images; but these methods are objective,and time-consuming.In this study,with a main data source of domestic high-resolution remote sensing images from ZY-3 satellite as well as the study area of the Wenchuan earthquake region,objects of multilevel landslides were established using the multi-scale optimum partition method based on in-depth analysis of landslide features.A recognition rule set of multi-dimensional landslides was also built through the combination of topographic features and image features,such as spectrum,texture,and geometry.Additionally,recognition models for landslide stratification were proposed based on the recognition models of high-resolution images and an understanding of the scenes.Through all of the aforementioned efforts,the spatial distribution of the seismic landslide as well as the sliding source area,transport area,and depositional area can be identified intelligently.The analysis results of the experimental area showed a minimum recognition accuracy of 82.97%,with the depositional zone of landslides being the easiest zone to recognize,and the effectiveness of the proposed method as well as ZY-3 data.These findings may provide technical support for regional earthquake-induced landslides investigations and further promote geological hazard application of domestic high-resolution satellites. 