Spatial Prediction of Landslide Hazard Using Logistic Regression and ROC Analysis

An empirical modeling of road related and non‐road related landslide hazard for a large geographical area using logistic regression in tandem with signal detection theory is presented. This modeling was developed using geographic information system (GIS) and remote sensing data, and was implemented on the Clearwater National Forest in central Idaho. The approach is based on explicit and quantitative environmental correlations between observed landslide occurrences, climate, parent material, and environmental attributes while the receiver operating characteristic (ROC) curves are used as a measure of performance of a predictive rule. The modeling results suggest that development of two independent models for road related and non‐road related landslide hazard was necessary because spatial prediction and predictor variables were different for these models. The probabilistic models of landslide potential may be used as a decision support tool in forest planning involving the maintenance, obliteration or development of new forest roads in steep mountainous terrain.     

GIS支持下应用PSO-SVM模型预测滑坡易发性

滑坡灾害易发性预测是滑坡监测、预警与评估的关键技术。如何有效地选取评价因子和构建预测模型是滑坡灾害定量预测研究中的难题。本文以三峡库区长江干流岸坡作为研究区,通过地形、地质和遥感等多源数据融合,提取滑坡孕灾环境和诱发因素的信息作为评价因子。在此基础上,针对滑坡灾害的非线性和不确定性特征,采用粒子群算法对支持向量机模型参数进行全局寻优,构建粒子群算法(particle swarm optimization, PSO)-支持向量机(support vector machine, SVM)模型,定量预测滑坡易发性。最后通过分类精度比较分析基于格网单元和对象单元的滑坡易发性预测精度,结果表明,基于对象单元的PSO-SVM预测精度较高,其曲线下面积为0.841 5,Kappa系数为0.849 0,预测结果与野外实际调查情况较为一致,可为三峡库区滑坡防灾减灾工作提供参考。     

模糊时间序列模型在滑坡预测预报中的应用

将模糊时间序列模型引入到滑坡预测预报中。简要介绍了模糊时间序列建模的步骤,并用实例验证了模糊时间序列模型用于滑坡预测预报的可行性。     

斜坡单元支持下的滑坡易发性评价支持向量机模型

针对传统滑坡预测手段数据源有限、数据更新周期长、难以发现隐藏在复杂滑坡系统中的规律等问题,本文以三峡库区为研究对象,从多源空间数据中提取滑坡孕灾环境和影响因素等信息,采用数字地形水文分析方法划分斜坡单元,对评价因子进行重采样,进而构建两类支持向量机模型。分析了多源影响因素与滑坡易发性的定量关系,并生成滑坡易发性分区图。采用成功率曲线和误差率评价预测结果,模型预测精度达到98.21%,与野外调查实际情况吻合较好。     

Spatial Interpolation Using Fuzzy Reasoning

Spatial interpolation is an important feature of a Geographic Information System, which is the procedure used to estimate values at unknown locations within the area covered by existing observations. In this paper, we describe a conservative spatial interpolation technique that incorporates the advantages of local interpolation, Euclidean interpolation, and conservative fuzzy reasoning, and a dynamic fuzzy–reasoning–based function estimator with parameters optimised by a genetic algorithm. The main objective of this paper is to formulate a computationally efficient spatial interpolation technique similar to the IDWA technique that can be used in real time application. The main feature of our spatial interpolation technique is a capability for spatial interpolation and extrapolation in a higher–dimensional space. Examples from a rainfall spatial interpolation problem are used to illustrate the applicability of the proposed technique.     

滑坡灾害遥感遥测预警理论及方法

本文系统地探讨了滑坡灾害多源遥感遥测数据综合建模预警理论及方法,体现了多学科交叉对滑坡灾害研究的贡献。主要内容如下：     

Assessment of Seismically Induced Landslide Hazard for the State of Karnataka Using GIS Technique

Landslide hazards are a major natural disaster that affects most of the hilly regions around the world. In India, significant damages due to earthquake induced landslides have been reported in the Himalayan region and also in the Western Ghat region. Thus there is a requirement of a quantitative macro-level landslide hazard assessment within the Indian subcontinent in order to identify the regions with high hazard. In the present study, the seismic landslide hazard for the entire state of Karnataka, India was assessed using topographic slope map, derived from the Digital Elevation Model (DEM) data. The available ASTER DEM data, resampled to 50 m resolution, was used for deriving the slope map of the entire state. Considering linear source model, deterministic seismic hazard analysis was carried out to estimate peak horizontal acceleration (PHA) at bedrock, for each of the grid points having terrain angle 10° and above. The surface level PHA was estimated using nonlinear site amplification technique, considering B-type NEHRP site class. Based on the surface level PHA and slope angle, the seismic landslide hazard for each grid point was estimated in terms of the static factor of safety required to resist landslide, using Newmark’s analysis. The analysis was carried out at the district level and the landslide hazard map for all the districts in the Karnataka state was developed first. These were then merged together to obtain a quantitative seismic landslide hazard map of the entire state of Karnataka. Spatial variations in the landslide hazard for all districts as well as for the entire state Karnataka is presented in this paper. The present study shows that the Western Ghat region of the Karnataka state is found to have high landslide hazard where the static factor of safety required to resist landslide is very high.     

GIS支持下滑坡灾害空间预测方法研究

滑坡预测在防灾减灾工作中具有重要意义,它包括空间、时间预测两个方面。基于统计模型进行区域评价与空间预测是滑坡灾害研究的重要方向,但是预测结果往往依赖样本数量和空间分布等。本文以马来西亚金马伦高原为研究区,选择高程、坡度、坡向、地表曲率、构造、土地覆盖、地貌类型、道路和排水系统作为评价因子,探讨运用地理信息系统(GIS)和遥感(RS)获取与管理滑坡灾害信息,以及热带雨林地区湿热环境下滑坡空间预测的方法。支持向量机(SVM)和逻辑(Logistic)回归模型分别应用于滑坡空间预测,结果显示平均预测精度分别为95.9%和86.2%,SVM法具有较高的描述精度,值得推荐;同时,基于SVM模型的滑坡空间预测受样本影响较小,预测结果相对比较稳定,这对于滑坡灾害区域评价与预测的快速实现具有实际意义。     

Spatial Prediction of Rainfall-Induced Landslides Using Aggregating One-Dependence Estimators Classifier

In this study, the spatial prediction of rainfall-induced landslides at the Pauri Gahwal area, Uttarakhand, India has been done using Aggregating One-Dependence Estimators (AODE) classifier which has not been applied earlier for landslide problems. Historical landslide locations have been collated with a set of influencing factors for landslide spatial analysis. The performance of the AODE model has been assessed using statistical analyzing methods and receiver operating characteristic curve technique. The predictive capability of the AODE model has also been compared with other popular landslide models namely Support Vector Machines (SVM), Radial Basis Function Neural Network (ANN-RBF), Logistic Regression (LR), and Naïve Bayes (NB). The result of analysis illustrates that the AODE model has highest predictability, followed by the SVM model, the ANN-RBF model, the LR model, and the NB model, respectively. Thus AODE is a promising method for the development of better landslide susceptibility map for proper landslide hazard management.     

基于GIS的西攀高速公路沿线滑坡灾害管理

滑坡是西攀高速公路主要的地质灾害之一,本文采用GIS技术建立了西攀高速公路沿线滑坡灾害管理系统,提高了滑坡灾害的管理效率。详细讨论了各种滑坡灾害专题制图和滑坡体三维可视化建模的过程和方法。这些是滑坡治理过程中重要的非工程措施,也为滑坡治理的工程措施的优化提供决策依据。     

滑坡变形分析与预报的新方法

对滑坡变形分析与预报的一些新方法如提升小波、模糊聚类和模糊神经网络等方法及应用进行了研究,为滑坡变形分析与预报提供了新的研究思路。     

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.     

基于核主成分分析和粒子群优化支持向量机的滑坡位移预测

利用核主成分分析法对滑坡位移影响因子进行特征提取,以获得的主成分作为支持向量机的特征向量建立支持向量机模型,其中模型参数通过粒子群算法进行选择优化,构建出核主成分分析和粒子群优化支持向量机协同模型,对滑坡相对位移进行预测。预测结果的平均绝对误差和相对误差分别为0.760和7.563%,与其他预测模型相比,其拟合和泛化能力最优,表明核主成分分析和粒子群优化支持向量机协同模型的预测结果与实际监测值具有很好的一致性。     

Landslide Hazard Assessment Using Random SubSpace Fuzzy Rules Based Classifier Ensemble and Probability Analysis of Rainfall Data: A Case Study at Mu Cang Chai District,Yen Bai Province (Viet Nam)

Landslide hazard assessment at the Mu Cang Chai district; Yen Bai province (Viet Nam) has been done using Random SubSpace fuzzy rules based Classifier Ensemble (RSSCE) method and probability analysis of rainfall data. RSSCE which is a novel classifier ensemble method has been applied to predict spatially landslide occurrences in the area. Prediction of temporally landslide occurrences in the present study has been done using rainfall data for the period 2008–2013. A total of fifteen landslide influencing factors namely slope, aspect, curvature, plan curvature, profile curvature, elevation, land use, lithology, rainfall, distance to faults, fault density, distance to roads, road density, distance to rivers, and river density have been utilized. The result of the analysis shows that RSSCE and probability analysis of rainfall data are promising methods for landslide hazard assessment. Finally, landslide hazard map has been generated by integrating spatial prediction and temporal probability analysis of landslides for the land use planning and landslide hazard management.     

基于半参数模型的滑坡段数据估计与预测

在滑坡段数据的参数平差模型的基础上,通过在模型中增加表示模型误差或系统误差的非参数分量,将半参数模型的估计理论应用到滑坡段参数估计与预测中,分离出数据中的系统误差,给出了滑坡稳定系数的预测公式;并以云阳-奉节长江顺层岸滑坡段实测数据为例,将半参数估计方法与最小二乘方法进行了比较分析,验证了半参数估计方法应用到估计和预测滑坡段数据的可行性和有效性。     

基于并行Tabu搜索和空间信息约束的遥感影像模糊聚类

在传统模糊C-均值聚类的基础上,引入了描述空间邻近关系的空间隶属度;采用Tabu搜索策略,抑制了模糊聚类的局部收敛性和对聚类中心初值的敏感性;提出了并行算法,有效地降低了影像分割的通信复杂度,提高了算法的搜索速度,实现了线性加速比。实验结果表明,改进算法有效地提高了聚类抗噪性能,减少了聚类迭代次数。     

An Evidential Reasoning Approach to Land Degradation Evaluation: Dempster-Shafer Theory of Evidence

This study explores the combination of expert and local knowledge for an integrated evaluation of land degradation in southern Mauritania. The expert knowledge uses two imprecise information and uncertainty management tools, Fuzzy measure theory and Dempster‐Shafer theory, mediated by expert knowledge, to integrate data collected using local knowledge. Fuzzy measures were found to be an efficient way of handling uncertainty of data compiled from local environmental knowledge and uniformly scaling all land degradation indicator variables to the same numerical range of values. The two techniques allowed scaling and integration of 13 evidential themes characterizing southern Mauritania. The cartographic outputs of this study show the potential of Fuzzy measure theory and Demspter‐Shafer theory as powerful complementary environmental resource management tools. On a scale of 0.0 to 1.0, the maps indicate the degree of degradation. The seriousness of land degradation decreases from settlements, roads, and watering points outward. As such, they constitute an excellent guide for resource allocation and rehablitation plans.     

重庆市滑坡灾害时空特征分析

重庆市是滑坡灾害高发的城市,滑坡灾害每年都会造成大量的人员伤亡和经济损失。本文结合历年滑坡统计资料,基于GIS平台,分析重庆市滑坡灾害的时空分布特征。自1980年起,重庆发生滑坡的次数明显增加,且集中发生在5~9月,尤其是7月。80%以上的滑坡为降雨引发,其次是自然灾害和人类活动,人类活动诱发滑坡灾害增长速度最快。重庆滑坡密度大的地区为市区以及万州区、忠县等三峡库区核心地带,滑坡高发地区有东北向西南方向变化的趋势。     

相空间重构理论支持下的滑坡预测方法

针对滑坡位移时间序列的非线性特性,引入基于相空间重构和最小二乘支持向量机(LSSVM)的预测法.利用Cao氏法确定嵌入维数,计算最佳延迟时间;在相空间中,利用LSSVM建立预测模型,以实例对滑坡进行计算,对LSSVM模型和BP神经网络模型进行了比较.结果表明:基于相空间重构和LSSVM的滑坡预测模型具有较高的精度,是科...