GIS and ANN model for landslide susceptibility mapping

Landslide hazard is as the probability of occurrence of a potentially damaging landslide phenomenon within specified period of time and within a given area. The susceptibility map provides the relative spatial probability of landslides occurrence. A study is presented of the application of GIS and artificial neural network model to landslide susceptibility mapping, with particular reference to landslides on natural terrain in this paper. The method has been applied to Lantau Island, the largest outlying island within the territory of Hong Kong. A three-level neural network model was constructed and trained by the back-propagate algorithm in the geographical database of the study area. The data in the database includes digital elevation modal and its derivatives, landslides distribution and their attributes, superficial geological maps, vegetation cover, the raingauges distribution and their 14 years 5-minute observation. Based on field inspection and analysis of correlation between terrain variables and landslides frequency, lithology, vegetation cover, slope gradient, slope aspect, slope curvature, elevation, the characteristic value, the rainstorms corresponding to the landslide, and distance to drainage line are considered to be related to landslide susceptibility in this study. The artificial neural network is then coupled with the ArcView3.2 GIS software to produce the landslide susceptibility map, which classifies the susceptibility into three levels: low, moderate, and high. The results from this study indicate that GIS coupled with artificial neural network model is a flexible and powerful approach to identify the spatial probability of hazards.     

GIS and ANN model for landslide susceptibility mapping

1 IntroductionThe population growth and the expansion of settlements and life-lines over hazardous areas exert increasingly great impact of natural disasters both in the developed and developing countries. In many countries, the economic losses and casualties due to landslides are greater than commonly recognized and generate a yearly loss of property larger than that from any other natural disasters, including earthquakes, floods and windstorms. Landslides in mountainous terrain often occur a…     

Landslide characteristics and slope instability modeling using GIS, Lantau Island, Hong Kong

Steep terrain and high a frequency of tropical rainstorms make landslide occurrence on natural terrain a common phenomenon in Hong Kong. This paper reports on the use of a Geographical Information Systems (GIS) database, compiled primarily from existing digital maps and aerial photographs, to describe the physical characteristics of landslides and the statistical relations of landslide frequency with the physical parameters contributing to the initiation of landslides on Lantau Island in Hong Kong. The horizontal travel length and the angle of reach, defined as the angle of the line connecting the head of the landslide source to the distal margin of the displaced mass, are used to describe runout behavior of landslide mass. For all landslides studied, the horizontal travel length of landslide mass ranges from 5 to 785 m, with a mean value of 43 m, and the average angle of reach is 27.7°. This GIS database is then used to obtain a logistic multiple regression model for predicting slope instability. It is indicated that slope gradient, lithology, elevation, slope aspect, and land-use are statistically significant in predicting slope instability, while slope morphology and proximity to drainage lines are not important and thus excluded from the model. This model is then imported back into the GIS to produce a map of predicted slope instability. The results of this study demonstrate that slope instability can be effectively modeled by using GIS technology and logistic multiple regression analysis.     

Spatial patterns of old, deep-seated landslides: A case-study in the northern Ethiopian highlands

During the last decade, slope failures were reported in a 500 km² study area in the Geba–Werei catchment, northern Ethiopia, a region where landslides were not considered an important hazard before. Field observations, however, revealed that many of the failures were actually reactivations of old deep-seated landslides after land use changes. Therefore, this study was conducted (1) to explore the importance of environmental factors controlling landslide occurrence and (2) to estimate future landslide susceptibility. A landslide inventory map of the study area derived from aerial photograph interpretation and field checks shows the location of 57 landslides and six zones with multiple landslides, mainly complex slides and debris flows. In total 14.8% of the area is affected by an old landslide. For the landslide susceptibility modelling, weights of evidence (WofE), was applied and five different models were produced. After comparison of the models and spatial validation using Receiver Operating Characteristic curves and Kappa values, a model combining data on elevation, hillslope gradient, aspect, geology and distance to faults was selected. This model confirmed our hypothesis that deep-seated landslides are located on hillslopes with a moderate slope gradient (i.e. 5°–13°). The depletion areas are expected on and along the border of plateaus where weathered basalts rich in smectite clays are found, and the landslide debris is expected to accumulate on the Amba Aradam sandstone and upper Antalo limestone. As future landslides are believed to occur on inherently unstable hillslopes similar to those where deep-seated landslides occurred, the classified landslide susceptibility map allows delineating zones where human interventions decreasing slope stability might cause slope failures. The results obtained demonstrate that the applied methodology could be used in similar areas where information on the location of landslides is essential for present-day hazard analysis.     

Predictive modelling of rainfall-induced landslide hazard in the Lesser Himalaya of Nepal based on weights-of-evidence

Landslide hazard mapping is a fundamental tool for disaster management activities in mountainous terrains. The main purpose of this study is to evaluate the predictive power of weights-of-evidence modelling in landslide hazard assessment in the Lesser Himalaya of Nepal. The modelling was performed within a geographical information system (GIS), to derive a landslide hazard map of the south-western marginal hills of the Kathmandu Valley. Thematic maps representing various factors (e.g., slope, aspect, relief, flow accumulation, distance to drainage, soil depth, engineering soil type, landuse, geology, distance to road and extreme one-day rainfall) that are related to landslide activity were generated, using field data and GIS techniques, at a scale of 1:10,000. Landslide events of the 1970s, 1980s, and 1990s were used to assess the Bayesian probability of landslides in each cell unit with respect to the causative factors. To assess the accuracy of the resulting landslide hazard map, it was correlated with a map of landslides triggered by the 2002 extreme rainfall events. The accuracy of the map was evaluated by various techniques, including the area under the curve, success rate and prediction rate. The resulting landslide hazard value calculated from the old landslide data showed a prediction accuracy of > 80%. The analysis suggests that geomorphological and human-related factors play significant roles in determining the probability value, while geological factors play only minor roles. Finally, after the rectification of the landslide hazard values of the new landslides using those of the old landslides, a landslide hazard map with > 88% prediction accuracy was prepared. The methodology appears to have extensive applicability to the Lesser Himalaya of Nepal, with the limitation that the model's performance is contingent on the availability of data from past landslides.     

汶川地震灾区滑坡空间特征变化分析

“5.12”汶川大地震触发了大量滑坡,给人民群众生命财产和社会经济发展造成了巨大损失。基于GIS空间分析方法,结合震前和震后的滑坡编目数据,对滑坡与坡度、坡向、高程、岩土类型、道路、河流和断裂带等7个孕灾环境因素的空间分布关系进行统计分析。结果表明：滑坡与孕灾环境因素的空间分布关系受地震的影响比较大。相比于震前,震后滑坡发生的优势坡度、优势岩土类型、优势距离缓冲区等均发生了很大的变化;并且坡向、距道路距离、距河流距离等因素对滑坡有明显地趋势性影响。     

汶川地震灾区滑坡空间特征变化分析简

“5.12”汶川大地震触发了大量滑坡,给人民群众生命财产和社会经济发展造成了巨大损失。基于GIS空间分析方法,结合震前和震后的滑坡编目数据,对滑坡与坡度、坡向、高程、岩土类型、道路、河流和断裂带等7个孕灾环境因素的空间分布关系进行统计分析。结果表明：滑坡与孕灾环境因素的空间分布关系受地震的影响比较大。相比于震前,震后滑坡发生的优势坡度、优势岩土类型、优势距离缓冲区等均发生了很大的变化;并且坡向、距道路距离、距河流距离等因素对滑坡有明显地趋势性影响。     

GIS支持下滑坡斜坡类型定量化及制图研究

斜坡类型描述岩层产状与斜坡的角度关系,很大程度上决定了斜坡岩土体变形的方式和强度,对地质灾害分布具有重要作用。斜坡的顺向坡、反向坡与地形的阳坡、阴坡概念相似,可以利用改进的太阳辐射地形因子计算模型(TOBIA指数)对斜坡类型进行定量化表达。计算TOBIA指数需要斜坡坡度、坡向、岩层倾角、倾向4个参数。以三峡库区顺向坡基岩滑坡多发地段青干河流域为例,通过区域地质图上产状点获取离散岩层倾角和倾向数值,经空间插值得到空间连续分布的倾角和倾向参数;通过数字高程模型获取坡度和坡向参数,得到区内TOBIA指数分布。在此基础上进一步研究指数和滑坡发育关系。结果表明,TOBIA指数值与区内斜坡类型密切相关,根据TOBIA指数值能很好地区分斜坡类型。以二分类变量逻辑回归模型对坡度和指数两个变量进行分析,发现引入TOBIA指数后,回归模型对已知滑坡拟合度由55%提高到71.5%,能有效提高区域滑坡灾害危险性区划结果精度。     

Probabilistic landslide hazard assessment at the basin scale

We propose a probabilistic model to determine landslide hazard at the basin scale. The model predicts where landslides will occur, how frequently they will occur, and how large they will be. We test the model in the Staffora River basin, in the northern Apennines, Italy. For the study area, we prepare a multi-temporal inventory map through the interpretation of multiple sets of aerial photographs taken between 1955 and 1999. We partition the basin into 2243 geo-morpho-hydrological units, and obtain the probability of spatial occurrence of landslides by discriminant analysis of thematic variables, including morphological, lithological, structural and land use. For each mapping unit, we obtain the landslide recurrence by dividing the total number of landslide events inventoried in the unit by the time span of the investigated period. Assuming that landslide recurrence will remain the same in the future, and adopting a Poisson probability model, we determine the exceedance probability of having one or more landslides in each mapping unit, for different periods. We obtain the probability of landslide size by analysing the frequency–area statistics of landslides, obtained from the multi-temporal inventory map. Assuming independence, we obtain a quantitative estimate of landslide hazard for each mapping unit as the joint probability of landslide size, of landslide temporal occurrence and of landslide spatial occurrence.     

Aspects of landslide activity in the Mercantour Massif and the French Riviera, southeastern France

Landslides pose serious hazards in the Mercantour Massif and the French Riviera in southeastern France. The context for landslide development is a particularly favourable one, both in terms of the geomorphic and structural setting of this Alpine region, and of the climatic, hydrologic and seismic factors that trigger such failures. High mountain relief and steep slopes constitute a very favourable setting for failures affecting massive basement rocks and a very heterogeneous sedimentary cover whose resistance has been weakened by weathering, tectonic stresses, and cambering due to gravity. Among trigger factors, the important appears to be the precipitation regime. Rainfalls are commonly concentrated into short high-intensity downpours or into bursts of sustained falls over periods of several days, leading to soil saturation and lubrication of potential failure planes. Snowmelt also contributes to these lubrication processes. Earthquakes affecting this region are also a potentially important landslide trigger. However, while a lot of work has been done on the relationship between extreme climatic events and landslide activity, much less is known of the trigger effects of earthquakes.Both the background factors that promote landslide development and the factors that trigger such failures are discussed within a time frame of landslide development. Progressive changes in soil strength due to weathering, rock cambering and shattering processes lead to long-term reduction in resistance. Superimposed on these progressive changes are episodic triggerings related to rainfall and snowmelt episodes or earthquakes. Landslides may occur as shallow, low-volume “one-time” events or may be part of a progressive long-term failure. The former generally affect unconsolidated or clay-rich sedimentary rocks, especially on the coastal hillslopes of the French Riviera. A notable exception of a major, voluminous “one-time” event was the submarine landslide of the Var Delta in 1979. This landslide, like numerous other smaller subaerial landslides onland, was largely a result of human activities. This landslide occurred following extensive modification of the Var Delta and, notably, reclamation of the steep, fine-grained delta front. Deforestation, quarrying, urbanisation and road network developments are various ways in which human activity has destabilized the coastal hillslopes, favouring the development of numerous shallow landslides following each episode of heavy rainfall.Progressive landslides on the upper hillslopes of the Mercantour Massif have developed over long time spans (order of 10¹ to 10⁵ yrs) and have involved more complex interactions between lithological controls, slope characteristics and trigger factors. The Collelongue and Bois de Malbosc landslides have evolved into now stable failures buttressed by resistant migmatitic diorites or amphibolites. The more voluminous and well monitored Clapière landslide is a relatively simple rotational landslide of the toe-failure type. This active landslide poses a serious to inhabitants and infrastructure in the Tinée Valley. The importance of continued field monitoring, modelling and mapping of landslides and their hazards is emphasised.     

Comparison of landslide susceptibility based on a decision-tree model and actual landslide occurrence: The Akaishi Mountains, Japan

This paper proposes a statistical decision-tree model to analyze landslide susceptibility in a wide area of the Akaishi Mountains, Japan. The objectives of this study were to validate the decision-tree model by comparing landslide susceptibility and actual landslide occurrence, and to reveal the relationships among landslide occurrence, topography, and geology. Landslide susceptibility was examined through ensemble learning with a decision tree. Decision trees are advantageous in that estimation processes and order of important explanatory variables are explicitly represented by the tree structures. Topographic characteristics (elevation, slope angle, profile curvature, plan curvature, and dissection and undissection height) and geological data were used as the explanatory variables. These topographic characteristics were calculated from digital elevation models (DEMs). The objective variables were landslide occurrence and reactivation data between 1992 and 2002 that were depicted by satellite image analysis. Landslide susceptibility was validated by comparing actual data on landslides that occurred and reactivated after the model was constructed (between 2002 and 2004).This study revealed that, from 2002 to 2004, landslides tended to occur and reactivate in catchments with high landslide susceptibility. The landslide susceptibility map thus depicts the actual landslide occurrence and reactivation in the Akaishi Mountains. This result indicates that the decision-tree model has appropriate accuracy for estimating the probabilities of future landslides. The tree structure indicates that landslides occurred and reactivated frequently in the catchments that had an average slope angle exceeding ca. 29° and a mode of slope angle exceeding 33°, which agree well with previous studies. A decision tree also quantitatively expresses important explanatory variables at the higher order of the tree structure.     

GIS analysis to assess landslide susceptibility in a fluvial basin of NW Sicily (Italy)

Landslide hazard assessment, effected by means of geostatistical methods, is based on the analysis of the relationships between landslides and the spatial distributions of some instability factors. Frequently such analyses are based on landslide inventories in which each record represents the entire unstable area and is managed as a single instability landform. In this research, landslide susceptibility is evaluated through the study of a variety of instability landforms: landslides, scarps and areas uphill from crown. The instability factors selected were: bedrock lithology, steepness, topographic wetness index and stream power index. The instability landform densities computed for all the factors, which were arranged in Unique Condition Unit, allowed us to derive a total of three prediction images for each landslide typology. The role of the instability factors and the effects generated by the use of different landforms were analyzed by means of: a) bivariate analysis of the relationships between factors and landslide density; b) predictive power validations of the prediction images, based on a random partition strategy.The test area was the Iato River Basin (North-Western Sicily), whose slopes are moderately involved in flow and rotational slide landslides (219 and 28, respectively). The area is mainly made up of the following complexes: Numidian Flysch clays (19%, 1%), Terravecchia sandy clays (5%, 1%), Terravecchia clayey sands (3%, 0.3%) and San Cipirello marly clays (9%, 0%). The steepness parameter shows the highest landslide density in the [11–19°] class for both the typologies (8%, 1%), even if the density distributions for rotational slides are right-asymmetric and right-shifted. We obtained significant differences in shape when we used different instability landforms. Unlike scarps and areas uphill from crowns, landslide areas produce left-asymmetric and left-shifted density distributions for both the typologies. As far as the topographic wetness index is concerned, much more pronounced differences were detected among the instability landforms of rotational slides. In contrast, the flow landslides produce normal-like density distributions. The latter and the rotational slide landslide areas produce the highest density values in the class [5.5–6.7], despite an abrupt decreasing trend starting from the first class [3.2–4.4], which is generated by the density values of the rotational slide scarps and areas uphill from crowns. The stream power index at the foot of the slopes, which was automatically derived using a GIS-procedure, shows a positive correlation with the landslide densities marked by the maximum classes: [4.8–6.0] for flows, and [6.0–7.2] for rotational slides. The validation procedure results confirmed that the choice of instability landform influences the results of the susceptibility analysis. Furthermore, the validation procedure indicates that: a) the predictive models are generally satisfactory; b) scarps and zones uphill from crown areas are the most diagnostically unstable landforms, for flow and rotational slide landslides respectively.     

Slope stability and landslides in proglacial varved clays of western Estonia

During the last decade the frequency of landslides at river valley slopes eroding into the glaciolacustrine plain in western Estonia has grown considerably. We studied in detail nine recent landslides out of 25 known and recorded sliding events in the area. All landslides occurred at the river banks in otherwise almost entirely flat areas of proglacial deposits capped with marine sands. Glaciolacustrine varved clay is the weakest soil type in the area and holds the largest landslides. Slope stability modelling shows that critical slope gradient for the clay is ≥ 10° and for the marine sand ≥ 20°. Fluvial erosion is the main process in decreasing slope stability at the outer bends of the river meanders. An extra shear stress generated by groundwater flow following the high stand of the groundwater level or rapid water level drawdown in the river channels are responsible for triggering the landslides. Consecutive occurrence of small-scale slides has a direct effect in triggering the large, retrogressive complexes of slides in the glaciolacustrine clay. A landslide hazard zonation map was composed based on digital elevation model and the data on spatial distribution of glaciolacustrine clays and marine sands, and on existing and critical slope angles of these deposits.     

Probability maps of landslide reactivation derived from tree-ring records: Pra Bellon landslide, southern French Alps

Probability maps of landslide reactivation are presented for the Pra Bellon landslide located in the southern French Alps based on results obtained with dendrogeomorphic analysis. Spatiotemporal patterns of past landslide activity was derived from tree-ring series of 403 disturbed mountain pine trees growing in the landslide body. In total, 704 growth disturbances were identified in the samples indicating 22 reactivation phases of the landslide body between 1910 and 2011. The mean return period was 4.5 years. Given the spatiotemporal completeness of the reconstruction, probabilities of landslide reactivation were computed and illustrated using a Poisson distribution model and for 5, 20, 50, and 100 years. Probability of landslide reactivation is highest in the central part of the landslide body and increases from 0.13 for a 5-year period to 0.94 for a 100-year period. Conversely, probabilities of reactivation are lower at its margins. The proposed method differs from conventional approaches based on statistical analyses or physical modeling that have demonstrated to have limitations in the prediction of spatiotemporal reactivation of landslides. Our approach is, in contrast, based on extensive data on past landslides and therefore allowed determination of quantitative probability maps of reactivation derived directly from the frequency of past events. This approach is considered a valuable tool for land managers in charge of protecting and forecasting people and their assets from the negative effects of landslides as well as for those responsible for land use planning and management. It demonstrates the reliability of dendrogeomorphic mapping that should be used systematically in forested shallow landslides.     

GIS支持下三峡库区秭归县滑坡灾害空间预测

基于GIS空间分析和统计模型相结合进行区域评价与空间预测是滑坡灾害研究的重要方向之一。以三峡库区秭归县为研究区,选择坡度、坡向、边坡结构、工程岩组、排水系统、土地利用和公路开挖作为评价因子。为提高模型的预测精度、可信度和推广能力,利用窗口采样规则降低训练样本之间的空间相关性。建立Logistic回归模型,对滑坡灾害与评价因子进行定量相关性分析。计算研究区滑坡灾害易发性指数,对其进行聚类分析,绘制滑坡易发性分区图,其中高、中易发区占整个研究区面积的38.9%,主要分布在人类工程活动频繁和靠近排水系统的区域。经过验证,该模型的预测精度达到77.57%。     

Landslides in the Central Coalfield (Cantabrian Mountains, NW Spain): Geomorphological features, conditioning factors and methodological implications in susceptibility assessment

A geomorphological study focussing on slope instability and landslide susceptibility modelling was performed on a 278 km² area in the Nalón River Basin (Central Coalfield, NW Spain). The methodology of the study includes: 1) geomorphological mapping at both 1:5000 and 1:25,000 scales based on air-photo interpretation and field work; 2) Digital Terrain Model (DTM) creation and overlay of geomorphological and DTM layers in a Geographical Information System (GIS); and 3) statistical treatment of variables using SPSS and development of a logistic regression model. A total of 603 mass movements including earth flow and debris flow were inventoried and were classified into two groups according to their size. This study focuses on the first group with small mass movements (10⁰ to 10¹ m in size), which often cause damage to infrastructures and even victims. The detected conditioning factors of these landslides are lithology (soils and colluviums), vegetation (pasture) and topography. DTM analyses show that high instabilities are linked to slopes with NE and SW orientations, curvature values between − 6 and − 0.7, and slope values from 16° to 30°. Bedrock lithology (Carboniferous sandstone and siltstone), presence of Quaternary soils and sediments, vegetation, and the topographical factors were used to develop a landslide susceptibility model using the logistic regression method. Application of “zoom method” allows us to accurately detect small mass movements using a 5-m grid cell data even if geomorphological mapping is done at a 1:25,000 scale.     

Spatial pattern and influencing factors of landslide casualty events

Analysis of casualties due to landslides from 2000 to 2012 revealed that their spatial pattern was affected by terrain and other natural environmental factors, which resulted in a higher distribution of landslide casualty events in southern China than in northern China. Hotspots of landslide-generated casualties were in the western Sichuan mountainous area and Yunnan-Guizhou Plateau region, southeast hilly area, northern part of the loess hilly area, and Tianshan and Qilian Mountains. However, local distribution patterns indicated that landslide casualty events were also influenced by economic activity factors. To quantitatively analyse the influence of natural environment and human-economic activity factors, the Probability Model for Landslide Casualty Events in China (LCEC) was built based on logistic regression analysis. The results showed that relative relief, GDP growth rate, mean annual precipitation, fault zones, and population density were positively correlated with casualties caused by landslides. Notably, GDP growth rate ranked only second to relative relief as the primary factors in the probability of casualties due to landslides. The occurrence probability of a landslide casualty event increased 2.706 times with a GDP growth rate increase of 2.72%. In contrast, vegetation coverage was negatively correlated with casualties caused by landslides. The LCEC model was then applied to calculate the occurrence probability of landslide casualty events for each county in China. The results showed that there are 27 counties with high occurrence probability but zero casualty events. The 27 counties were divided into three categories: poverty-stricken counties, mineral-rich counties, and real-estate overexploited counties; these are key areas that should be emphasized in reducing landslide risk.     

Predicting landslides for risk analysis — Spatial models tested by a cross-validation technique

A landslide-hazard map is intended to show the location of future slope instability. Most spatial models of the hazard lack reliability tests of the procedures and predictions for estimating the probabilities of future landslides, thus precluding use of the maps for probabilistic risk analysis. To correct this deficiency we propose a systematic procedure comprising two analytical steps: “relative-hazard mapping” and “empirical probability estimation”. A mathematical model first generates a prediction map by dividing an area into “prediction” classes according to the relative likelihood of occurrence of future landslides, conditional by local geomorphic and topographic characteristics. The second stage estimates empirically the probability of landslide occurrence in each prediction class, by applying a cross-validation technique. Cross-validation, a “blind test” here using non-overlapping spatial or temporal subsets of mapped landslides, evaluates accuracy of the prediction and from the resulting statistics estimates occurrence probabilities of future landslides. This quantitative approach, exemplified by several experiments in an area near Lisbon, Portugal, can accommodate any subsequent analysis of landslide risk.     

中国地质灾害伤亡事件的空间格局及影响因素（英文）

Analysis of casualties due to landslides from 2000 to 2012 revealed that their spatial pattern was affected by terrain and other natural environmental factors, which resulted in a higher distribution of landslide casualty events in southern China than in northern China. Hotspots of landslide-generated casualties were in the western Sichuan mountainous area and Yunnan-Guizhou Plateau region, southeast hilly area, northern part of the loess hilly area, and Tianshan and Qilian Mountains. However, local distribution patterns indicated that landslide casualty events were also influenced by economic activity factors. To quantitatively analyse the influence of natural environment and human-economic activity factors, the Probability Model for Landslide Casualty Events in China(LCEC) was built based on logistic regression analysis. The results showed that relative relief, GDP growth rate, mean annual precipitation, fault zones, and population density were positively correlated with casualties caused by landslides. Notably, GDP growth rate ranked only second to relative relief as the primary factors in the probability of casualties due to landslides. The occurrence probability of a landslide casualty event increased 2.706 times with a GDP growth rate increase of 2.72%. In contrast, vegetation coverage was negatively correlated with casualties caused by landslides. The LCEC model was then applied to calculate the occurrence probability of landslide casualty events for each county in China. The results showed that there are 27 counties with high occurrence probability but zero casualty events. The 27 counties were divided into three categories: poverty-stricken counties, mineral-rich counties, and real-estate overexploited counties; these are key areas that should be emphasized in reducing landslide risk.