降雨型滑坡时间概率的逻辑回归拟合及连续概率滑坡危险性建模

提高降雨型滑坡危险性预警精度和空间辨识度具有重要意义.以江西宁都县1980—2001年156个降雨型滑坡为例,首先基于传统的EE-D（early effective rainfall-rainfall duration）阈值法计算不同降雨诱发滑坡的时间概率级别;然后以各级别临界降雨阈值曲线对应的时间概率为因变量,并以对应的前期有效降雨量（early effective rainfall,EE）和降雨历时（D）为自变量,采用逻辑回归拟合出上述因变量与自变量之间的非线性关系,得到降雨诱发滑坡的连续概率值;之后对比C5.0决策树和多层感知器的滑坡易发性预测性能;最后利用降雨诱发滑坡的连续概率值与易发性图相耦合以实现连续概率滑坡危险性预警.结果显示:（1）宁都降雨型滑坡连续概率值的逻辑回归方程为1/P=1+e4.062+0.747 4×D-0.079 44×EE,其拟合优度为0.983;（2）2002—2003年的20处用于连续概率阈值测试的降雨型滑坡大都落在连续概率值大于0.7的区域,只有4处落在小于0.7的区域;（3）C5.0决策树预测滑坡易发性的精度显著高于多层感知器;（4）近5年的4次降雨型滑坡的连续概率危险性值都在0.8以上,且高和极高预警区的面积较传统滑坡危险性分区更小.可见连续概率滑坡危险性预警法相较于传统危险性分区法具有更高的预警精度和空间辨识度,且通过叠加滑坡易发性图及其临界降雨阈值可开展实时滑坡危险性预警制图.      

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.      

基于易发性分区的区域滑坡降雨预警阈值确定——以云南龙陵县为例

滑坡所处不同易发等级的区域,降雨预警阈值差别较大。为提高滑坡降雨预警的针对性和准确率,文章以野外地质调查和滑坡易发条件分析为基础,结合信息量模型和层次分析法开展滑坡易发性评价,再通过滑坡发生概率与前期累计降雨量的相关性分析,分区进行滑坡降雨预警阈值模型研究。结果表明：坡度、高程、距断层距离、工程地质岩组、水系是龙陵县滑坡的主要孕灾地质条件;龙陵县滑坡非易发区面积为14.33 km²,低易发区面积为1 053.87 km²,中易发区面积为1 471.65 km²,高易发区面积为254.73 km²;确定单日和前3日为降雨预警时间,分区分时细化了降雨预警阈值模型;对比降雨预警阈值模型应用于龙陵县滑坡监测预警中的前后,预警信息减少了70条,预警准确率提高了14.4%,并实现了镇安镇户帕村施家寨组滑坡的有效预警。文章为区域滑坡降雨预警阈值确定提供了一种较好的参考方法。     

降雨型黄土滑坡预警研究现状综述

黄土滑坡是一类典型的地质灾害,但目前并没有针对这类滑坡的区域性预警模型。介绍了黄土滑坡的分类和滑带特征研究现状,讨论了降雨型滑坡的两类预警模型:降雨型滑坡临界雨量模型,包括区域统计概率预警和临界雨量阈值预警;物理预警模型,主要为通过建立力学稳定性分析和降雨入渗模型耦合得到的预警模型。分析了降雨或地下水作用下土体含水量变化对滑坡稳定性的影响研究和黄土滑坡的预警研究现状,最后根据前人研究现状讨论了现有预警模型的特点并提出了未来降雨型黄土滑坡预警可能的发展方向。     

降雨型滑坡定量风险评价研究——以宣恩县干坝滑坡为例

单体滑坡定量风险评价一直是滑坡研究领域的重点及难点,特别是滑坡危险程度和易损性的确定。本文以湖北宣恩县干坝滑坡为例,进行降雨型滑坡定量风险评价研究;在降雨极值分析的基础上,计算不同降雨重现期下滑坡渗流场和稳定性变化规律;在滑坡影响范围分析的基础上,采用定性方法确定不同区域承灾体的易损性,定量分析不同降雨重现期下逐个建筑物和室内人员风险,通过建立人口和财产总风险与年超越概率关系曲线计算滑坡年风险;通过三维数字模型展示建筑物和室内人员在不同降雨工况下的风险,利于主管部门对单体滑坡的风险管控。     

降雨滑坡预警的概率分析方法

滑坡启动的降雨临界值是滑坡预警的关键,由于研究区资料数据有限,关键值难以确定,降雨滑坡预警效果受影响。所以应该对预警结果进行概率分析。本文利用作者在已开发的降雨滑坡预警系统,提出采用降雨滑坡预警概率分析方法进行预警。通过对滑坡与雨量相关性、降雨滑坡启动值等对滑坡预警的时间概率、空间概率及预警概率进行分析,计算危险区内已发生滑坡频率和降雨滑坡发生频率,得到降雨滑坡预警概率使得预警系统更加可靠。并以沐川县为例对模型进行了算例检验,为更科学的对降雨滑坡进行预警提供方法借鉴。     

基于机器学习的滑坡崩塌地质灾害气象风险预警研究

在划分气象风险等级时,传统地质灾害气象风险预警方法忽略了承灾体脆弱性因素,且气象风险预报等级整体偏高,导致高等级风险区空报率较高。基于此,提出基于机器学习的滑坡、崩塌灾害气象风险预警方法。利用信息量法,分析气象因素影响程度。选取坐标点、降雨量、易发生等级,将其作为机器学习人工神经网络的输入节点,判断是否发生崩塌、滑坡灾害;针对地质灾害区域,根据影响程度计算气象引发因子指数,结合滑坡、崩塌灾害潜势度G和承灾体脆弱性M,确定气象风险预警指数R,划分预警级别,完成滑坡、崩塌灾害气象风险预警。实验结果表明,设计方法有效降低了三级预报和四级预警空报率,提升了预警精细化程度。     

湖南茶陵滑坡空间预警的降雨临界值初步分析

湖南省茶陵县位于湖南东部,滑坡是县内分布最广、危害最大的地质灾害类型,且有土质、浅层、规模小、降雨诱发的特点。本文以茶陵县已知发灾时间的滑坡作为样本,建立滑坡灾害的发生与降雨量的统计关系,确定导致茶陵县不同区域滑坡发生的临界降雨量,最后利用地质-气象耦合的预报模式得出茶陵县滑坡灾害气象预警的分级和分区。     

降雨诱发区域性滑坡预报预警方法研究

论文以四川省雅安市雨城区为研究区，将逻辑回归模型引入区域降雨型滑坡预警预报，建立了同时考虑降雨强度和降雨过程的降雨临界值表达式。在滑坡危险性区划的基础上，研究提出了区域降雨型滑坡预警预报指标，包括临界值降雨指数R和滑坡发生指数L，并利用20台自动遥测雨量计和地质灾害群测群防网络，采用历史记录雨量和预报雨量，建立了区域降雨型滑坡预报预警体系。     

基于有效降雨强度和逻辑回归的降雨型 滑坡预测模型

以湖南省张家界市桑植县为研究区,在全面分析近30年降雨及滑坡数据的基础上,对滑坡及滑坡数量与降雨因子的关系开展了统计分析研究。首先确定了区域最佳有效降雨衰减系数,同时分别按滑坡规模、坡度、厚度大小统计了降雨与历史滑坡信息,得出有效降雨强度（I）与持续时间（D）散点图,由此确定各不同概率下诱发滑坡的区域有效降雨强度阈值,并进行了滑坡灾害危险性等级划分。进而,利用部分样本数据进行逻辑回归分析,得到了该研究区的滑坡发生概率预测方程,并给出了降雨强度临界值定量表达式,最后选用实际降雨诱发滑坡事件与未诱发滑坡事件进行对比验证。结果表明,文章所建立的滑坡预测模型准确性较高,预测情况与实际情况比较吻合。     

降雨引发的兰州黄土滑坡时空规律分析和临界降雨量预测

开展降雨型黄土滑坡预警对于区域性防治滑坡具有重要意义。本研究在收集1985~2015年兰州市降雨型黄土滑坡历史数据的基础上,运用反距离权重插值(IDW)和核密度估算(KDE)方法揭示了降雨引发黄土滑坡的时空分布规律。该文基于统计学的基本原理,运用相关性和偏相关性等方法建立适合兰州市的有效降雨量模型。通过拟合有效降雨量与滑坡因子的线性回归关系,确定引发黄土滑坡的临界降雨量阈值,设定兰州市黄土滑坡的降雨量危险性预警等级。研究表明:(1)兰州市黄土滑坡灾害点沿着黄河及其支流沿岸分布,城关区滑坡点最多且呈环形分布,西固区次之,其他地区分布较少;(2)降雨是兰州市及其周边地区黄土滑坡的关键诱因,10d有效降雨量与滑坡因子均呈现显著正相关特性,其相关系数达到0.698;(3)依据10mm、20mm和40mm临界降雨量阈值将预警等级划分为低、中、高3个危险性等级。     

Regional prediction of landslide hazard using probability analysis of intense rainfall in the Hoa Binh province, Vietnam

The main objective of this study is to assess regional landslide hazards in the Hoa Binh province of Vietnam. A landslide inventory map was constructed from various sources with data mainly for a period of 21 years from 1990 to 2010. The historic inventory of these failures shows that rainfall is the main triggering factor in this region. The probability of the occurrence of episodes of rainfall and the rainfall threshold were deduced from records of rainfall for the aforementioned period. The rainfall threshold model was generated based on daily and cumulative values of antecedent rainfall of the landslide events. The result shows that 15-day antecedent rainfall gives the best fit for the existing landslides in the inventory. The rainfall threshold model was validated using the rainfall and landslide events that occurred in 2010 that were not considered in building the threshold model. The result was used for estimating temporal probability of a landslide to occur using a Poisson probability model. Prior to this work, five landslide susceptibility maps were constructed for the study area using support vector machines, logistic regression, evidential belief functions, Bayesian-regularized neural networks, and neuro-fuzzy models. These susceptibility maps provide information on the spatial prediction probability of landslide occurrence in the area. Finally, landslide hazard maps were generated by integrating the spatial and the temporal probability of landslide. A total of 15 specific landslide hazard maps were generated considering three time periods of 1, 3, and 5 years.     

A review of assessing landslide frequency for hazard zoning purposes

The probability of occurrence is one of the key components of the risk equation. To assess this probability in landslide risk analysis, two different approaches have been traditionally used. In the first one, the occurrence of landslides is obtained by computing the probability of failure of a slope (or the reactivation of existing landslides). In the second one, which is the objective of this paper, the probability is obtained by means of the statistical analysis of past landslide events, specifically by the assessment of the past landslide frequency. In its turn, the temporal frequency of landslides may be determined based on the occurrence of landslides or from the recurrence of the landslide triggering events over a regional extent. Hazard assessment using frequency of landslides, which may be taken either individually or collectively, requires complete records of landslide events, which is difficult in some areas. Its main advantage is that it may be easily implemented for zoning. Frequency assessed from the recurrence of landslide triggers, does not require landslide series but it is necessary to establish reliable relations between the trigger, its magnitude and the occurrence of the landslides. The frequency of the landslide triggers can be directly used for landslide zoning. However, because it does not provide information on the spatial distribution of the potential landslides, it has to be combined with landslide susceptibility (spatial probability analysis) to perform landslide hazard zoning. Both the scale of work and availability of data affect the results of the landslide frequency and restrict the spatial resolution of frequency zoning as well. Magnitude–frequency relationships are fundamental elements for the quantitative assessment of both hazard and risk.     

基于逻辑回归的四川青川县区域滑坡灾害预警模型

四川省青川县滑坡灾害群发,点多面广,区域滑坡灾害预警是有效防灾减灾的重要手段,预警模型是成功预警的核心。由于研究区滑坡诱发机理复杂、调查监测大数据及分析方法不足等原因,传统区域地质灾害预警模型存在预警精度有限、精细化不足等问题。文章在青川县地质灾害调查监测和降水监测成果集成整理与数据清洗基础上,构建了青川县区域滑坡灾害训练样本集,样本集包括地质环境、降雨等27个输入特征属性和1个输出特征属性,涵盖了青川县近9年（2010—2018年）全部样本,数量达1 826个（其中,正样本613个,负样本1 213个）。基于逻辑回归算法,对样本集进行5折交叉验证学习训练,采用贝叶斯优化算法进行模型优化,采用精确度、ROC曲线和AUC值等指标校验模型准确度和模型泛化能力。其中,ROC曲线也称为“受试者工作特征”曲线;AUC值表示ROC曲线下的面积。校验结果显示,基于逻辑回归算法的模型训练结果准确率和泛化能力均较好（准确率94.3%,AUC为0.980）。开展区域滑坡实际预警时,按训练样本特征属性格式,输入研究区各预警单元27个特征属性,调用预先学习训练好的模型,输出滑坡灾害发生概率,根据输出概率分段确定滑坡灾害预警等级。当输出概率P≥40%且P<60%时,发布黄色预警;当输出概率P≥60%且P<80%时,发布橙色预警;当输出概率P≥80%时,发布红色预警。     

Early warning of rainfall-induced shallow landslides and debris flows in the USA

The state of knowledge and resources available to issue alerts of precipitation-induced landslides vary across the USA. Federal and state agencies currently issue warnings of the potential for shallow, rapidly moving landslides and debris flows in a few areas along the Pacific coast and for areas affected by Atlantic hurricanes. However, these agencies generally lack resources needed to provide continuous support or to expand services to other areas. Precipitation thresholds that form the basis of landslide warning systems now exist for a few areas of the USA, but the threshold rainfall amounts and durations vary over three orders of magnitude nationwide and over an order of magnitude across small geographic areas such as a county. Antecedent moisture conditions also have a significant effect, particularly in areas that have distinct wet and dry seasons. Early warnings of shallow landslides that include specific information about affected areas, probability of landslide occurrence, and expected timing are technically feasible as illustrated by a case study from the Seattle, WA area. The four-level warning scheme (Null, Outlook, Watch, Warning) defined for Seattle is based on observed or predicted exceedance of a cumulative precipitation threshold and a rainfall intensity–duration threshold combined with real-time monitoring of soil moisture. Based on analysis of historical data, threshold performance varies according to precipitation characteristics, and threshold exceedance corresponds to a given probability of landslide occurrence. Experience in Seattle during December 2004 and January 2005 illustrates some of the challenges of providing landslide early warning on the USA West Coast.     

有限元法分析降雨条件下斜坡稳定性——以邻水县马鞍坪滑坡为例

通过对华蓥市马鞍坪滑坡现有变形监测情况表明,斜坡变形位移与降雨的关系十分密切,马鞍坪滑坡的后部强变形区,斜坡位移与50mm以上持续降雨几乎呈对应关系。研究人员采用工程地质条件分析与有限元分析法相结合,对马鞍坪滑坡进行了滑坡体在降雨条件下饱和一非饱和渗流分析与坡体稳定性分析。分析结果表明,降雨量是华蓥山地区斜坡失稳下滑的主要引发因素,不同雨量决定斜坡演化成滑坡的不同阶段,其影响作用也不近相同。计算结果表明,当日降雨量达到150mm左右且累积降雨量超过300mm的情况下,斜坡土体将出现变形,出现滑坡可能性大。该分析结果为今后华蓥山地区地质灾害监测预警示范中的地质灾害单体预警预报,提供了可靠预警参数。     

基于台风路径追踪的滑坡概率分析

台风暴雨是我国东南丘陵山地滑坡的主要诱发因素,揭示台风路径与滑坡发生的相关关系对东南丘陵山地台风暴雨型土质滑坡监测预警具有重要的理论及实际意义。本文基于2015—2019年直接过境福建省或间接对福建省造成影响的台风数据以及与这些台风事件发生期间的降雨量数据和台风暴雨滑坡数据,运用ArcGIS软件中的克里金插值法将台风路径、降雨量数据及台风暴雨滑坡数据进行耦合。再运用Logistic回归方法,通过SPSS软件得到泉州市基于台风路径追踪的滑坡发生概率模型。并运用克里金插值法和Logistic回归方法,以台风杜鹃期间的滑坡为实例对所得模型进行验证,其对实际雨量站测得降雨数据引发滑坡与否的判对概率为77%,对实际发生滑坡的判对概率为100%,依据实际降雨量计算数据,其误报率为21%,但实际滑坡发生的误报率为0,符合安全性。预报效果较为满意,因此模型可作为在台风经过泉州市时的台风暴雨型滑坡发生概率的预测。     

Erratum to: Early warning of rainfall-induced shallow landslides and debris flows in the USA

The state of knowledge and resources available to issue alerts of precipitation-induced landslides vary across the USA. Federal and state agencies currently issue warnings of the potential for shallow, rapidly moving landslides and debris flows in a few areas along the Pacific coast and for areas affected by Atlantic hurricanes. However, these agencies generally lack resources needed to provide continuous support or to expand services to other areas. Precipitation thresholds that form the basis of landslide warning systems now exist for a few areas of the USA, but the threshold rainfall amounts and durations vary over three orders of magnitude nationwide and over an order of magnitude across small geographic areas such as a county. Antecedent moisture conditions also have a significant effect, particularly in areas that have distinct wet and dry seasons. Early warnings of shallow landslides that include specific information about affected areas, probability of landslide occurrence, and expected timing are technically feasible as illustrated by a case study from the Seattle, WA area. The four-level warning scheme (Null, Outlook, Watch, Warning) defined for Seattle is based on observed or predicted exceedance of a cumulative precipitation threshold and a rainfall intensity–duration threshold combined with real-time monitoring of soil moisture. Based on analysis of historical data, threshold performance varies according to precipitation characteristics, and threshold exceedance corresponds to a given probability of landslide occurrence. Experience in Seattle during December 2004 and January 2005 illustrates some of the challenges of providing landslide early warning on the USA West Coast.     

Quantitative assessment on the influence of heavy rainfall on the crystalline schist landslide by monitoring system -case study on Zentoku landslide, Japan

On Shikoku Island, which is one of the four main islands of Japan, a large number of large-scale crystalline schist landslides have been revealed and are being monitored by an observation system. Seasonal heavy rainfall is the most active meteorological factor that can threaten the stability of this kind of site-specific landslide. In this paper, on the basis of the study of the rainfall-related behavior of a typical crystalline schist landslide, the Zentoku landslide, by analyzing the precisely and continuously observed piezometric and movement data, a method was developed to quantitatively assess the effect of heavy rainfall on a large-scale landslide. The results indicated that heavy rainfall-induced landslide displacement shows good correlation with the variation of groundwater levels. Variations of groundwater level have been simulated with the use of a tank model. The simulation using this model permits the change in water levels for future rainfall events to be predicted. By combining the predicted results with the empirical relation between displacements and water levels, rainfall-induced landslide movement during extreme rainfall events can be estimated in advance. The effect of heavy rainfall on sliding behavior can be quantified in terms of the change in displacement. Thus warning information or advisories for the local residents can be provided.