基于Mamdani FIS的中吉乌铁路方案线中段滑坡易发性预测

中吉乌铁路是我国西北地区通往中亚、南欧国家的一条国际通路,对其方案线沿线的地质灾害的调查及预测可为其选线提供一定建议。本文基于Mamdani模糊推理系统(Mamdani FIS)对方案线北线AK53-AK130、南线AK61-AK131段研究区的滑坡易发性进行预测。通过区域地质背景资料和遥感影像人机交互解译获取了该区地质环境背景、地形因素以及生态环境3类9种滑坡影响因子,建立768条推理规则,通过Mamdani FIS模型得到区内滑坡敏感度文件,在GIS环境中制作研究区滑坡易发性等级图,将研究区划分为极低易发区、低易发区、中等易发区、高易发区和极高易发区。结果显示,滑坡极高易发区和高易发区分布在研究区东北部的费尔干纳山脉附近以及南部的亚瑟河流域。采用受试者工作特征曲线(Receiver Operating Characteristic Curve,ROC曲线)对结果进行验证,曲线下面积为0.859,表明预测结果精度较高。     

基于遥感影像多尺度分割与地质因子评价的滑坡易发性区划

区域滑坡易发性的研究是滑坡空间预测的核心内容之一。从影像多尺度分割和面向对象的分类理论出发,以研究区遥感影像的熵、能量、相关性、对比度共4个参数作为影像纹理因子提取易发性特征,利用滑坡所处区域的库水影响等级、坡度、斜坡结构、工程岩组4类地质因子分析地质背景,搭建C5.0决策树的易发性分类模型,实现了对研究区内4类滑坡易发性单元的预测。结果表明:高易发性单元的工程岩组通常发育为软岩岩组和软硬相间岩组,且坡度在15°～30°之间;模型显示该区域训练样本和测试样本平均正确率达91.64%,Kappa系数分别为0.84,0.51,因此这种基于影像多尺度分割与地质因子分级的滑坡易发性分类研究具有一定的适用性。     

几种聚类优化的机器学习方法在灵台县滑坡易发性评价中的应用

笔者以甘肃省平凉市灵台县为目标研究区域,基于地理空间和历史滑坡数据,利用混合高斯聚类(GMM)优化的逻辑回归(LR)、支持向量机(SVM)、BP神经网络(BP Neural Network)、随机森林(RF)4种机器学习模型构建滑坡易发性评价分析模型。选取高程、坡度、坡向、曲率、黄土侵蚀强度、归一化植被指数、地质构造7个环境因子作为滑坡易发性影响因子,以30m栅格建立影响因子地理空间数据库,将研究区域划分为180万栅格单元。利用混合高斯聚类模型对整个研究区域的栅格单元进行聚类,得出初步的滑坡易发分区,选择易发程度最低类别中的栅格单元作为非滑坡区域,每次随机选择500个单元作为非滑坡单元,并根据历史滑坡数据将203个已知滑坡栅格单元作为滑坡单元,建立4种机器学习分类模型。利用训练好的模型对整个研究区域进行预测,绘制各算法的受试者工作曲线(ROC曲线),对各个算法的预测结果进行对比。分析结果表明,在本目标研究区域,各模型的滑坡易发区划图与实际的滑坡分布情况总体相吻合。随机森林模型的ROC曲线下面积(AUC)最大为0.96,测试集准确率最高为0.93;BP神经网络模型的ROC曲线下面积和测试集准确率次之,为0.90和0.87;支持向量机模型和逻辑回归模型的ROC曲线下面积和测试集准确率分别为0.86、0.81和0.85、0.80,均低于随机森林和BP神经网络模型。     

基于GIS与ANN模型的地震滑坡易发性区划

基于遥感数据、地理信息系统(GIS)技术和人工神经网络(ANN)模型,开展地震滑坡易发性区划研究.2010年4月14日玉树地震后,基于航片与卫星影像目视解译,并辅以野外调查的方法,在地震区圈定了2036处地震诱发滑坡.选择高程、坡度、坡向、斜坡曲率、坡位、与水系距离、地层岩性、与断裂距离、与公路距离、归一化植被指数(NDVI)、与同震地表破裂距离、地震动峰值加速度(PGA)共12个因子作为地震滑坡易发性评价因子.这些因子均是应用GIS技术与遥感影像处理技术,基于地形数据、地质数据、遥感数据得到.训练样本中的滑动样本有两组,一组是滑坡区整个单滑坡体的质心位置,另一组是滑坡滑源区滑前的坡体高程最高的位置.应用这12个影响因子,分别采用这两组评价样本,基于ANN模型建立地震滑坡易发性索引图,基于GIS工具建立地震滑坡易发性分级图.分别应用训练样本中滑坡分布的点数据去检验各自的结果正确率,正确率分别为81.53％与81.29％,表明ANN模型是一种高效科学的地震滑坡易发性区划模型.     

基于C5.0决策树-快速聚类模型的万州区库岸段乡镇滑坡易发性区划

以乡镇为评价单元开展区域滑坡易发性评价对用地规划、防灾减灾等方面具有重要意义。以万州区临江段的23个乡镇单元作为研究对象,首先选取地表高程、坡度、坡向、岩性、构造、土地利用类型、地形湿度指数、水系、道路9个指标因子,通过C5.0决策树算法计算该区域发生滑坡的概率,再利用快速聚类算法进行易发性结果分级;基于ArcGIS平台得到各乡镇单元的滑坡易发性分区,结果表明:C5.0决策树-快速聚类模型的易发性评价精度最高,AUC值达到0.950,优于人工神经网络-快速聚类模型的0.826和贝叶斯-快速聚类模型的0.772。利用C5.0决策树-快速聚类模型的计算结果,综合考虑极高(高)易发区面积大小及其占乡镇面积比大小,完成各乡镇单元的滑坡易发性区划。在所有23个乡镇中,滑坡易发性等级高的包括大周镇、万州城区、溪口乡、新田镇等乡镇。通过对比各乡镇滑坡面积占研究区滑坡总面积的比重,发现两者结论基本一致,预测结果可为全区滑坡防灾减灾提供科学依据。     

基于逻辑回归模型和3S技术的思南县滑坡易发性评价

区域滑坡易发性评价对滑坡灾害防治具有重要意义,贵州省思南县由于其特殊的自然地理和地质条件,受滑坡地质灾害的影响非常严重,因此,非常有必要对思南县的滑坡易发性进行评价。在滑坡编录的基础上,采用由RS、GIS和GPS组成的3S技术,获取了思南县的数字高程模型、坡度、坡向、剖面曲率、坡长、岩土类型、地表湿度指数、距离水系的距离、植被覆盖度和地表建筑物指数10个滑坡影响因子;再在频率比和相关性分析的基础上,利用逻辑回归模型对思南县的滑坡易发性进行了评价并绘制了易发性分布图。结果表明:利用逻辑回归模型预测思南县滑坡易发性的准确率(AUC值)达到0.797,较为准确地预测出了思南县滑坡分布规律;极高和高滑坡易发区主要分布在高程低于600 m、地表坡度较大且以软质岩类为主的区域;而极低和低滑坡易发区主要分布在高程较高、地表坡度较小且以硬质岩类为主的区域。     

基于信息量、逻辑回归及其耦合模型的滑坡易发性评估研究:以青海沙塘川流域为例

滑坡易发性定量评估是预测滑坡发生空间概率的重要手段,基于统计分析原理的评估方法目前在国内外应用最为广泛,且不同评估方法的对比研究逐渐成为热点。以青海沙塘川流域黄土梁峁区为例,剖析了信息量模型和逻辑回归模型在滑坡易发性评估中的优越性和局限性,并探索提出基于二者的耦合模型。考虑坡度、坡向、起伏度、岩性、与干流距离、与支流距离和植被指数等7个影响因素,对比分析了基于信息量、逻辑回归及二者耦合模型的滑坡易发性评估的技术流程及结果。3种模型的成功率分别为:耦合模型成功率(78. 9%)>信息量模型成功率(71. 8%)>逻辑回归模型成功率(70. 8%)。在沙塘川流域黄土滑坡的易发性评估中,信息量和逻辑回归模型的表现基本相当,但信息量-逻辑回归耦合模型的成功率明显提升。该研究结果可为黄土高原区滑坡易发性定量评估提供借鉴。     

不同环境因子联接和预测模型的滑坡易发性建模不确定性

拟深入探讨滑坡与其环境因子间的非线性联接计算以及不同数据驱动模型等因素,对滑坡易发性预测建模不确定性的影响规律.以江西省瑞金市为例共获取370处滑坡和10种环境因子,通过概率统计（probability statistics,PS）、频率比（frequency ratio,FR）、信息量（information value,Ⅳ）、熵指数（index of entropy,IOE）和证据权（weight of evidence,WOE）等5种联接方法分别耦合逻辑回归（logistic regression,LR）、BP神经网络（BP neural networks,BPNN）、支持向量机（support vector machines,SVM）和随机森林（random forest,RF）模型共构建出20种耦合模型,同时构建无联接方法直接将原始数据作为输入变量的4种单独LR、BPNN、SVM和RF模型,预测出总计24种工况下的滑坡易发性;最后分别使用ROC曲线、均值、标准差和差异显著性等指标分析上述24种工况下易发性结果的不确定性.结果表明:（1）基于WOE的耦合模型预测滑坡易发性的平均精度最高且不确定性较低,基于PS的耦合模型预测精度最低且不确定性最高,基于FR、Ⅳ和IOE的耦合模型介于两者之间;（2）单独数据驱动模型易发性预测精度略低于耦合模型,且未能计算出环境因子各子区间对滑坡发育的影响规律,但其建模效率高于耦合模型;（3）RF模型预测精度最高且不确定性较低,其次分别为SVM、BPNN和LR模型.总之WOE是更优秀的联接法且RF模型预测性能最优,WOE-RF模型预测的滑坡易发性不确定性较低且更符合实际滑坡概率分布特征.      

人工神经网络在贵州地质灾害易发性分析中的应用探索

贵州省地质环境条件复杂、地质灾害发育严重,开展基于人工神经网络算法的地质灾害易发性分析,对提高全省易发性评价效率、实现智能化分析具有重要意义。本文通过对各种人工神经网络算法的比较,选取径向基神经网络(RBF)、概率神经网络(PNN)、模糊神经网络(FNN)作为模型算法,采用基于GIS平台的空间数据建模软件(SDM)构建基于人工智能的地质灾害风险分析模型系统,通过数据准备、数据处理、模型训练、模型调用、评估优化等步骤,开展基于三种人工神经网络的贵州山区地质灾害易发性分析应用探索。结果表明:(1)三种人工神经网络计算的易发性结果分区合理、精度检验合格,AUC检验显示具有良好预测价值;(2)通过与专家经验分析结果对比,RBF算法的评价结果与实际更为吻合,表明RBF算法可更好地应用于地质灾害易发性分析。     

基于有效降雨强度的滑坡灾害危险性预警

选取湖北省恩施地区1 000 km2区域作为典型研究区, 在全面分析该区域历史滑坡资料的基础上, 根据该区滑坡生成与地层岩性之间的关系, 将研究区地层划分为高、中、低3类易发性岩组.分岩组统计降雨监测数据与历史滑坡信息, 得出有效降雨强度与关键降雨持续时间的散点图, 由此确定不同滑坡发生概率的有效降雨强度阈值, 提出该区的滑坡灾害危险性预警判别模型.基于样本区统计数据建立滑坡预测指标体系, 运用GIS得出研究区域的滑坡空间易发性区划结果, 并根据不同易发岩组-有效降雨强度模型, 叠加滑坡灾害易发性分区结果与降雨危险性预警等级分级结果, 对研究区的滑坡灾害危险性进行了预测预警.结果表明: 不同易发岩组-有效降雨强度模型所得预警结果与实际情况吻合, 预警模型具有考虑全面和预警精度高的特点, 在实际预警中切实可用.      

空间三维滑坡敏感性分区工具及其应用

对于滑坡敏感性分区目前有三种方法:定性法、统计法和基于岩土定量模型的确定性方法。定性法基于对滑坡敏感性或灾害评估的人为判断;统计法用一个来源于结合了权重因子的预测函数或指标;而确定性法,或者说是物理定量模型法以质量、能量和动量守恒定律为基础。二维确定性模型广泛用于土木工程设计,而无限边坡模型(一维)也用于滑坡灾害分区的确定性模型。文中提出了一个新的基于GIS(地理信息系统)的滑坡敏感性分区系统,这个系统可用于从复杂地形中确认可能的危险三维(3-D)滑坡体。所有与滑坡相关的空间数据(矢量或栅格数据)都被集成到这个系统中。通过把研究区域划分为边坡单元并假定初始滑动面是椭球的下半部分,并使用Monte Carlo随机搜索法,三维滑坡稳定性分析中的三维最危险滑面是三维安全系数最小的地方。使用近似方法假定有效凝聚力、有效摩擦角和三维安全系数服从正态分布,可以计算出滑坡失稳概率。3DSlopeGIS是一个计算机程序,它内嵌了GIS Developer kit(ArcObjects of ESRI)来实现GIS空间分析功能和有效的数据管理。应用此工具可以解决所有的三维边坡空间数据解问题。通过使用空间分析、数据管理和GIS的可视化功能来处理复杂的边坡数据,三维边坡稳定性问题很容易用一个友好的可视化图形界面来解决。将3DSlopeGIS系统应用到3个滑坡敏感性分区的实例中:第一个是一个城市规划项目,第二个是预测以往滑坡灾害对临近区域可能的影响,第三个则是沿着国家主干道的滑坡分区。基于足够次数的Monte Carlo模拟法,可以确认可能的最危险滑坡体。这在以往的传统边坡稳定性分析中是不可能的。     

Landslide susceptibility modeling based on ANFIS with teaching-learning-based optimization and Satin bowerbird optimizer

As threats of landslide hazards have become gradually more severe in recent decades,studies on landslide prevention and mitigation have attracted widespread attention in relevant domains.A hot research topic has been the ability to predict landslide susceptibility,which can be used to design schemes of land exploitation and urban development in mountainous areas.In this study,the teaching-learning-based optimization(TLBO)and satin bowerbird optimizer(SBO)algorithms were applied to optimize the adaptive neuro-fuzzy inference system(ANFIS)model for landslide susceptibility mapping.In the study area,152 landslides were identified and randomly divided into two groups as training(70%)and validation(30%)dataset.Additionally,a total of fifteen landslide influencing factors were selected.The relative importance and weights of various influencing factors were determined using the step-wise weight assessment ratio analysis(SWARA)method.Finally,the comprehensive performance of the two models was validated and compared using various indexes,such as the root mean square error(RMSE),processing time,convergence,and area under receiver operating characteristic curves(AUROC).The results demonstrated that the AUROC values of the ANFIS,ANFIS-TLBO and ANFIS-SBO models with the training data were 0.808,0.785 and 0.755,respectively.In terms of the validation dataset,the ANFISSBO model exhibited a higher AUROC value of 0.781,while the AUROC value of the ANFIS-TLBO and ANFIS models were 0.749 and 0.681,respectively.Moreover,the ANFIS-SBO model showed lower RMSE values for the validation dataset,indicating that the SBO algorithm had a better optimization capability.Meanwhile,the processing time and convergence of the ANFIS-SBO model were far superior to those of the ANFIS-TLBO model.Therefore,both the ensemble models proposed in this paper can generate adequate results,and the ANFIS-SBO model is recommended as the more suitable model for landslide susceptibility assessment in the study area considered due to its excellent accuracy and efficiency.     

Landslides susceptibility mapping based on geographical information system,GuiZhou, south-west China

The purpose of this study is to assess the susceptibility of landslides around the area of Guizhou province, in south-west of China, using a geographical information system (GIS). The base map is prepared by visiting the field area and mapping individual landslide at a scale of 1:500,000 topographic maps. In the study, slope, lithology, landslide inventory, tectonic activity, drainage distribution and annual precipitation were taken as independent causal factors. Therefore, six causal factors maps are prepared by collecting information from various authorized sources and converting them in to GIS maps. The susceptibility assessment is based on the qualitative map combination model and trapezoidal fuzzy number weighting (TFNW) approach. Using a predicted map of probability, the study area was classified into four categories of landslide susceptibility: low, moderate, high and very high. In addition, the weighting procedure showed that the TFNW is an efficient method for landslide causal factors weighting.     

Application of expert rules in indirect approaches for landslide susceptibility assessment

Landslide susceptibility (LS) assessment by indirect approaches presents some limitations due to (1) the tendency to simplify the environmental factors (i.e., variables) and (2) the assumptions that landslides occur under the same combination of variables for a study site. Recently, some authors have discussed the interest to introduce expert knowledge in the indirect approaches in order to improve the quality of indirect LS maps. However, if the results are reliable, the procedures used seem fastidious and a very good knowledge of the study site is essential. The objectives of this paper are to discuss a methodology to introduce the expert knowledge in the indirect mapping process. After the definition of the expert rules associated to three landslide types, several indirect LS maps are produced by two indirect exploratory approaches, based on fuzzy set theory and on a modification of a bivariate method called expert weight of evidence. Then, the indirect LS maps are confronted to a landslide inventory and a LS map produced by a direct approach. The analyses indicate that the methodology used to introduce the expert rules in the mapping process increases the predictive power of indirect LS map. Finally, some indications about advantages and drawbacks of each approach are given to help the geoscientist to introduce his expert knowledge in the landslide susceptibility mapping process.     

Geology,geodynamics and orogenic gold prospectivity modelling of the Paleoproterozoic Kumasi Basin,Ghana, West Africa

This paper describes the geology and tectonics of the Paleoproterozoic Kumasi Basin, Ghana, West Africa, as applied to predictive mapping of prospectivity for orogenic gold mineral systems within the basin. The main objective of the study was to identify the most prospective ground for orogenic gold deposits within the Paleoproterozoic Kumasi Basin. A knowledge-driven, two-stage fuzzy inference system (FIS) was used for prospectivity modelling. The spatial proxies that served as input to the FIS were derived based on a conceptual model of gold mineral systems in the Kumasi Basin. As a first step, key components of the mineral system were predictively modelled using a Mamdani-type FIS. The second step involved combining the individual FIS outputs using a conjunction (product) operator to produce a continuous-scale prospectivity map. Using a cumulative area fuzzy favourability (CAFF) curve approach, this map was reclassified into a ternary prospectivity map divided into high-prospectivity, moderate-prospectivity and low-prospectivity areas, respectively. The spatial distribution of the known gold deposits within the study area relative to that of the prospective and non-prospective areas served as a means for evaluating the capture efficiency of our model. Approximately 99% of the known gold deposits and occurrences fall within high- and moderate-prospectivity areas that occupy 31% of the total study area. The high- and moderate-prospectivity areas illustrated by the prospectivity map are elongate features that are spatially coincident with areas of structural complexity along and reactivation during D4 of NE–SW-striking D2 thrust faults and subsidiary structures, implying a strong structural control on gold mineralization in the Kumasi Basin. In conclusion, our FIS approach to mapping gold prospectivity, which was based entirely on the conceptual reasoning of expert geologists and ignored the spatial distribution of known gold deposits for prospectivity estimation, effectively captured the main mineralized trends. As such, this study also demonstrates the effectiveness of FIS in capturing the linguistic reasoning of expert knowledge by exploration geologists. In spite of using a large number of variables, the curse of dimensionality was precluded because no training data are required for parameter estimation.     

GIS-based landslide susceptibility mapping using bivariate statistical analysis in Devrek (Zonguldak-Turkey)

Devrek town with increasing population is located in a hillslope area where some landslides exist. Therefore, landslide susceptibility map of the area is required. The purpose of this study was to generate a landslide susceptibility map using a bivariate statistical index and evaluate and compare the results of the statistical analysis conducted with three different approaches in seed cell concept resulting in different data sets in Geographical Information Systems (GIS) based landslide susceptibility mapping applied to the Devrek region. The data sets are created from the seed cells of (a) crowns and flanks, (b) only crowns, and (c) only flanks of the landslides by using ten different causative parameters of the study area. To increase the data dependency of the analysis, all parameter maps are classified into equal frequency classes based directly on the percentile divisions of each corresponding seed cell data set. The resultant maps of the landslide susceptibility analysis indicate that all data sets produce fairly acceptable results. In each data set analysis, elevation, lithology, slope, aspect, and drainage density parameters are found to be the most contributing factors in landslide occurrences. The results of the three data sets are compared using Seed Cell Area Indexes (SCAI). This comparison shows that the crown data set produces the most accurate and successful landslide susceptibility map of the study area.     

Predictive landslide susceptibility mapping using spatial information in the Pechabun area of Thailand

For predictive landslide susceptibility mapping, this study applied and verified probability model, the frequency ratio and statistical model, logistic regression at Pechabun, Thailand, using a geographic information system (GIS) and remote sensing. Landslide locations were identified in the study area from interpretation of aerial photographs and field surveys, and maps of the topography, geology and land cover were constructed to spatial database. The factors that influence landslide occurrence, such as slope gradient, slope aspect and curvature of topography and distance from drainage were calculated from the topographic database. Lithology and distance from fault were extracted and calculated from the geology database. Land cover was classified from Landsat TM satellite image. The frequency ratio and logistic regression coefficient were overlaid for landslide susceptibility mapping as each factor’s ratings. Then the landslide susceptibility map was verified and compared using the existing landslide location. As the verification results, the frequency ratio model showed 76.39% and logistic regression model showed 70.42% in prediction accuracy. The method can be used to reduce hazards associated with landslides and to plan land cover.     

Comparing data-driven landslide susceptibility models based on participatory landslide inventory mapping in Purwosari area,Yogyakarta, Java

There are different approaches and techniques for landslide susceptibility mapping. However, no agreement has been reached in both the procedure and the use of specific controlling factors employed in the landslide susceptibility mapping. Each model has its own assumption, and the result may differ from place to place. Different landslide controlling factors and the completeness of landslide inventory may also affect the different result. Incomplete landslide inventory may produce significance error in the interpretation of the relationship between landslide and controlling factor. Comparing landslide susceptibility models using complete inventory is essential in order to identify the most realistic landslide susceptibility approach applied typically in the tropical region Indonesia. Purwosari area, Java, which has total 182 landslides occurred from 1979 to 2011, was selected as study area to evaluate three data-driven landslide susceptibility models, i.e., weight of evidence, logistic regression, and artificial neural network. Landslide in the study area is usually affected by rainfall and anthropogenic activities. The landslide typology consists of shallow translational and rotational slide. The elevation, slope, aspect, plan curvature, profile curvature, stream power index, topographic wetness index, distance to river, land use, and distance to road were selected as landslide controlling factors for the analysis. Considering the accuracy and the precision evaluations, the weight of evidence represents considerably the most realistic prediction capacities (79%) when comparing with the logistic regression (72%) and artificial neural network (71%). The linear model shows more powerful result than the nonlinear models because it fits to the area where complete landslide inventory is available, the landscape is not varied, and the occurence of landslide is evenly distributed to the class of controlling factor.