Landslide susceptibility mapping using certainty factor, index of entropy and logistic regression models in GIS and their comparison at Mugling�CNarayanghat road section in Nepal Himalaya

Landslide susceptibility maps are vital for disaster management and for planning development activities in the mountainous country like Nepal. In the present study, landslide susceptibility assessment of Mugling?CNarayanghat road and its surrounding area is made using bivariate (certainty factor and index of entropy) and multivariate (logistic regression) models. At first, a landslide inventory map was prepared using earlier reports and aerial photographs as well as by carrying out field survey. As a result, 321 landslides were mapped and out of which 241 (75?%) were randomly selected for building landslide susceptibility models, while the remaining 80 (25?%) were used for validating the models. The effectiveness of landslide susceptibility assessment using GIS and statistics is based on appropriate selection of the factors which play a dominant role in slope stability. In this case study, the following landslide conditioning factors were evaluated: slope gradient; slope aspect; altitude; plan curvature; lithology; land use; distance from faults, rivers and roads; topographic wetness index; stream power index; and sediment transport index. These factors were prepared from topographic map, drainage map, road map, and the geological map. Finally, the validation of landslide susceptibility map was carried out using receiver operating characteristic (ROC) curves. The ROC plot estimation results showed that the susceptibility map using index of entropy model with AUC value of 0.9016 has highest prediction accuracy of 90.16?%. Similarly, the susceptibility maps produced using logistic regression model and certainty factor model showed 86.29 and 83.57?% of prediction accuracy, respectively. Furthermore, the ROC plot showed that the success rate of all the three models performed more than 80?% accuracy (i.e. 89.15?% for IOE model, 89.10?% for LR model and 87.21?% for CF model). Hence, it is concluded that all the models employed in this study showed reasonably good accuracy in predicting the landslide susceptibility of Mugling?CNarayanghat road section. These landslide susceptibility maps can be used for preliminary land use planning and hazard mitigation purpose.     

The 2010 Yushu earthquake triggered landslide hazard mapping using GIS and weight of evidence modeling

The Yushu County, Qinghai Province, China, April 14, 2010, earthquake triggered thousands of landslides in a zone between 96°20′32.9″E and 97°10′8.9″E, and 32°52′6.7″N and 33°19′47.9″N. This study examines the use of geographic information system (GIS) technology and Bayesian statistics in creating a suitable landslide hazard-zone map of good predictive power. A total of 2,036 landslides were interpreted from high-resolution aerial photographs and multi-source satellite images pre- and post-earthquake, and verified by selected field checking before a final landslide-inventory map of the study area could be established using GIS software. The 2,036 landslides were randomly partitioned into two subsets: a training dataset, which contains 80 % (1,628 landslides), for training the model; and a testing dataset 20 % (408 landslides). Twelve earthquake triggered landslide associated controlling parameters, such as elevation, slope gradient, slope aspect, slope curvature, topographic position, distance from main surface ruptures, peak ground acceleration, distance from roads, normalized difference vegetation index, distance from drainages, lithology, and distance from all faults were obtained from variety of data sources. Landslide hazard indices were calculated using the weight of evidence model. The landslide hazard map was compared with training data and testing data to obtain the success rate and predictive rate of the model, respectively. The validation results showed satisfactory agreement between the hazard map and the existing landslide distribution data. The success rate is 80.607 %, and the predictive rate is 78.855 %. The resulting landslide hazard map showed five classes of landslide hazard, i.e., very high, high, moderate, low and very low. The landslide hazard evaluation map should be useful for environmental recovery planning and reconstruction work.     

Calibration and validation of a model of forest disturbance in the Western Ghats,India 1920–1990

This paper introduces a new statistical method that we recommend should become standard procedure to quantify the goodness-of-fit of calibration and validation for land-use change models. We present a multiple-resolution Relative Operating Characteristic (ROC) that measures the goodness-of-fit between a reference map, which is considered reality, and a simulated map, which is a model’s output. This proposed ROC is based on: (1) multiple-resolutions, (2) soft classification, (3) sampling without replacement, and 4) explicit separation of factors of quantity versus location of land-use change. We illustrate the method with a case study in India’s Western Ghats, a biodiversity hotspot, where we have maps of cumulative forest disturbance for 1920 and 1990. We use a predictive modeling approach similar to GEOMOD in which we calibrate the model with the map of 1920, then predict the map of 1990, at which point we subject the model to validation. We show that the fit of calibration tends to be much larger than the fit of validation. Thus if a modeler assess a model by goodness-of-fit of calibration only, then the modeler will likely be over confident in the model’s predictive ability.     

基于人工神经网络模型的福建南平市滑坡危险性评价

滑坡灾害持续影响着人民生命财产安全和地区社会经济可持续发展,滑坡危险性评价能够为防灾减灾和区域规划提供有效的理论依据。以福建省南平市为研究区,区内1711个历史滑坡灾害点,选择高程、坡度、坡向、曲率、地质岩性、土壤类型、降雨、水系、土地利用类型、公路和铁路共11个影响因子构成基本评价体系。使用Spearman相关系数对各因子进行共线性分析。基于1711个滑坡样本和1711个随机选取的非滑坡样本数据,利用人工神经网络模型对研究区进行了滑坡危险性评价,并利用混淆矩阵和接收者操作特征曲线（ROC）对模型进行验证。结果表明:混淆矩阵精度84.91%,ROC曲线下面积AUC值0.93,说明模型具有较高精度和预测率。使用自然间断法将滑坡危险性分为5个等级,结果表明研究区内危险性最高地区位于延平区和浦城县,顺昌县和松溪县次之,其余地区多为低危险区和较低危险区。研究结果可为当地区域规划和防灾减灾工程提供一定的理论依据和科学指导。     

A comparative study of landslide susceptibility mapping using weight of evidence,logistic regression and support vector machine and evaluated by SBAS-InSAR monitoring: Zhouqu to Wudu segment in Bailong River Basin,China

The determining of landslide-prone areas in mountainous terrain is essential for land planning and hazard mitigation. In this paper, a comparative study using three statistical models including weight of evidence model (WoE), logistic regression model (LR) and support vector machine method (SVM) was undertaken in the Zhouqu to Wudu segment in the Bailong River Basin, Southern Gansu, China. Six conditionally independent environmental factors, elevation, slope, aspect, distance from fault, lithology and settlement density, were selected as the explanatory variables that may contribute to landslide occurrence based on principal component analysis (PCA) and Chi-square test. The relation between landslide distributions and these variables was analyzed using the three models and the results then used to calculate the landslide susceptibility (LS). The performance of the models was then evaluated using both the highly accurate deformation signals produced by using the Small Baseline Subset Interferometric Synthetic Aperture Radar technique and Receiver Operating Characteristic (ROC) curve. Results show more deformation points in areas with high and very high LS levels, and also more stable points in areas with low and very low LS levels for the SVM model. In addition, the SVM has larger area under the ROC curve. It indicates that the SVM has better prediction accuracy and classified ability. For the interpretability, the WoE derives the class of factors that most contributed to landsliding in the study area, and the LR reveals that factors including elevation, settlement density and distance from fault played major roles in landslide occurrence and distribution, whereas the SVM cannot provide relative weights for the variables. The outperformed SVM could be employed to determine potential landslide zones in the study area. Outcome of this research would provide preliminary basis for general land planning such as choosing new urban areas and infrastructure construction in the future, as well as for landslide hazard mitigation in Bailong River Basin.     

Flood risk analyses—how detailed do we need to be?

Applied flood risk analyses, especially in urban areas, very often pose the question how detailed the analysis needs to be in order to give a realistic figure of the expected risk. The methods used in research and practical applications range from very basic approaches with numerous simplifying assumptions up to very sophisticated, data and calculation time demanding applications both on the hazard and on the vulnerability part of the risk. In order to shed some light on the question of required model complexity in flood risk analyses and outputs sufficiently fulfilling the task at hand, a number of combinations of models of different complexity both on the hazard and on the vulnerability side were tested in a case study. The different models can be organized in a model matrix of different complexity levels: On the hazard side, the approaches/models selected were (A) linear interpolation of gauge water levels and intersection with a digital elevation model (DEM), (B) a mixed 1D/2D hydraulic model with simplifying assumptions (LISFLOOD-FP) and (C) a Saint-Venant 2D zero-inertia hyperbolic hydraulic model considering the built environment and infrastructure. On the vulnerability side, the models used for the estimation of direct damage to residential buildings are in order of increasing complexity: (I) meso-scale stage-damage functions applied to CORINE land cover data, (II) the rule-based meso-scale model FLEMOps+ using census data on the municipal building stock and CORINE land cover data and (III) a rule-based micro-scale model applied to a detailed building inventory. Besides the inundation depths, the latter two models consider different building types and qualities as well as the level of private precaution and contamination of the floodwater. The models were applied in a municipality in east Germany, Eilenburg. It suffered extraordinary damage during the flood of August 2002, which was well documented as were the inundation extent and depths. These data provide an almost unique data set for the validation of flood risk analyses. The analysis shows that the combination of the 1D/2D model and the meso-scale damage model FLEMOps+ performed best and provide the best compromise between data requirements, simulation effort, and an acceptable accuracy of the results. The more detailed approaches suffered from complex model set-up, high data requirements, and long computation times.     

A GIS-based approach for gully erosion susceptibility modelling: a test in Sicily, Italy

The aim of this study is to analyze the susceptibility conditions to gully erosion phenomena in the Magazzolo River basin and to test a method that allows for driving the factors selection. The study area is one of the largest (225 km²) watershed of southern Sicily and it is mostly characterized by gentle slopes carved into clayey and evaporitic sediments, except for the northern sector where carbonatic rocks give rise to steep slopes. In order to obtain a quantitative evaluation of gully erosion susceptibility, statistical relationships between the spatial distributions of gullies affecting the area and a set of twelve environmental variables were analyzed. Stereoscopic analysis of aerial photographs dated 2000, and field surveys carried out in 2006, allowed us to map about a thousand landforms produced by linear water erosion processes, classifiable as ephemeral and permanent gullies. The linear density of the gullies, computed on each of the factors classes, was assumed as the function expressing the susceptibility level of the latter. A 40-m digital elevation model (DEM) prepared from 1:10,000-scale topographic maps was used to compute the values of nine topographic attributes (primary: slope, aspect, plan curvature, profile curvature, general curvature, tangential curvature; secondary: stream power index; topographic wetness index; LS-USLE factor); from available thematic maps and field checks three other physical attributes (lithology, soil texture, land use) were derived. For each of these variables, a 40-m grid layer was generated, reclassifying the topographic variables according to their standard deviation values. In order to evaluate the controlling role of the selected predictive variables, one-variable susceptibility models, based on the spatial relationships between each single factor and gullies, were produced and submitted to a validation procedure. The latter was carried out by evaluating the predictive performance of models trained on one half of the landform archive and tested on the other. Large differences of accuracy were verified by computing geometric indexes of the validation curves (prediction and success rate curves; ROC curves) drawn for each one-variable model; in particular, soil texture, general curvature and aspect demonstrated a weak or a null influence on the spatial distribution of gullies within the studied area, while, on the contrary, tangential curvature, stream power index and plan curvature showed high predictive skills. Hence, predictive models were produced on a multi-variable basis, by variously combining the one-variable models. The validation of the multi-variables models, which generally indicated quite satisfactory results, were used as a sensitivity analysis tool to evaluate differences in the prediction results produced by changing the set of combined physical attributes. The sensitivity analysis pointed out that by increasing the number of combined environmental variables, an improvement of the susceptibility assessment is produced; this is true with the exception of adding to the multi-variables models a variable, as slope aspect, not correlated to the target variable. The addition of this attribute produces effects on the validation curves that are not distinguishable from noise and, as a consequence, the slope aspect was excluded from the final multi-variables model used to draw the gully erosion susceptibility map of the Magazzolo River basin. In conclusion, the research showed that the validation of one-variable models can be used as a tool for selecting factors to be combined to prepare the best performing multi-variables gully erosion susceptibility model.     

信息量模型、确定性系数模型与逻辑回归模型组合评价地质灾害敏感性的对比研究

为探索区域地质灾害敏感性评价方法,以宁夏盐池县为研究区域,选取坡度、坡向、坡高、高程、地层、距河流距离、距道路距离、植被覆盖度等8个影响地质灾害发生的评价因子,分别采用信息量模型+逻辑回归模型(I+LR)和确定性系数模型+逻辑回归模型(CF+LR)2种组合模型对盐池县地质灾害敏感性进行评价,将该区域地质灾害划分为极低、低、中和高敏感区4类,并完成结果检验。结果表明：(1)2种组合模型得到的低、中敏感区面积基本相当,而高敏感区面积相差较大,CF+LR模型较I+LR模型高敏感区面积增加约5336 km2,而极低敏感区面积减少约6%;(2)2种组合模型的合理性均符合检验要求,且ROC精度检验AUC值分别为0868和0829,渐进Sigb均小于005,表明2种组合评价模型都能较为客观准确地评价盐池县地质灾害敏感性;(3)ROC检验精度与盐池县地质灾害发育情况均表明I+LR模型精度更高。     

Statistical testing of a new testate amoeba‐based transfer function for water‐table depth reconstruction on ombrotrophic peatlands in north‐eastern Canada and Maine,United States

Proxy reconstructions of climatic parameters developed using transfer functions are central to the testing of many palaeoclimatic hypotheses on Holocene timescales. However, recent work shows that the mathematical models underpinning many existing transfer functions are susceptible to spatial autocorrelation, clustered training set design and the uneven sampling of environmental gradients. This may result in over‐optimistic performance statistics or, in extreme cases, a lack of predictive power. A new testate amoeba‐based transfer function is presented that fully incorporates the new recommended statistical tests to address these issues. Leave‐one‐out cross‐validation, the most commonly applied method in recent studies to assess model performance, produced over‐optimistic performance statistics for all models tested. However, the preferred model, developed using weighted averaging with tolerance downweighting, retained a predictive capacity equivalent to other published models even when less optimistic performance statistics were chosen. Application of the new statistical tests in the development of transfer functions provides a more thorough assessment of performance and greater confidence in reconstructions based on them. Only when the wider research community have sufficient confidence in transfer function‐based proxy reconstructions will they be commonly used in data comparison and palaeoclimate modelling studies of broader scientific relevance. Copyright © 2012 John Wiley & Sons, Ltd.     

The application of GIS-based bivariate statistical methods for landslide hazards assessment in the upper Tons river valley,Western Himalaya,India

Landslides are one of the most frequent and common natural hazards in many parts of Himalaya. To reduce the potential risk, the landslide susceptibility maps are one of the first and most important steps in the landslide hazard mitigation. Earth observation satellite and geographical information system-based techniques have been used to derive and analyse various geo-environmental parameters significant to landslide hazards. In this study, a bivariate statistics method was used for spatial modelling of landslide susceptibility zones. For this purpose, thematic layers including landslide inventory, geology, slope angle, slope aspect, geomorphology, slope morphology, drainage density, lineament and land use/land cover were used. A large number of landslide occurrences have been observed in the upper Tons river valley area of Western Himalaya. The result has been used to spatially classify the study area into zones of very high, high, moderate, low and very low landslide susceptibility zones. About 72% of active landslides have been observed to occur in very high and high hazard zones. The result of the analysis was verified using the landslide location data. The validation result shows significant agreement between the susceptibility map and landslide location. The result can be used to reduce landslide hazards by proper planning.     

Comparison of machine learning models for gully erosion susceptibility mapping

Gully erosion is a disruptive phenomenon which extensively affects the Iranian territory,especially in the Northern provinces.A number of studies have been recently undertaken to study this process and to predict it over space and ultimately,in a broader national effort,to limit its negative effects on local communities.We focused on the Bastam watershed where 9.3% of its surface is currently affected by gullying.Machine learning algorithms are currently under the magnifying glass across the geomorphological community for their high predictive ability.However,unlike the bivariate statistical models,their structure does not provide intuitive and quantifiable measures of environmental preconditioning factors.To cope with such weakness,we interpret preconditioning causes on the basis of a bivariate approach namely,Index of Entropy.And,we performed the susceptibility mapping procedure by testing three extensions of a decision tree model namely,Alternating Decision Tree(ADTree),Naive-Bayes tree(NBTree),and Logistic Model Tree(LMT).We dichotomized the gully information over space into gully presence/absence conditions,which we further explored in their calibration and validation stages.Being the presence/absence information and associated factors identical,the resulting differences are only due to the algorithmic structures of the three models we chose.Such differences are not significant in terms of performances;in fact,the three models produce outstanding predictive AUC measures(ADTree=0.922;NBTree=0.939;LMT=0.944).However,the associated mapping results depict very different patterns where only the LMT is associated with reasonable susceptibility patterns.This is a strong indication of what model combines best performance and mapping for any natural hazard-oriented application.     

Expected impacts of climate change on extreme climate events

An overview of the expected change of climate extremes during this century due to greenhouse gases and aerosol anthropogenic emissions is presented. The most commonly used methodologies rely on the dynamical or statistical downscaling of climate projections, performed with coupled atmosphere–ocean general circulation models. Either of dynamical or of statistical type, downscaling methods present strengths and weaknesses, but neither their validation on present climate conditions, nor their potential ability to project the impact of climate change on extreme event statistics allows one to give a specific advantage to one of the two types. The results synthesized in the last IPCC report and more recent studies underline a convergence for a very likely increase in heat wave episodes over land surfaces, linked to the mean warming and the increase in temperature variability. In addition, the number of days of frost should decrease and the growing season length should increase. The projected increase in heavy precipitation events appears also as very likely over most areas and also seems linked to a change in the shape of the precipitation intensity distribution. The global trends for drought duration are less consistent between models and downscaling methodologies, due to their regional variability. The change of wind-related extremes is also regionally dependent, and associated to a poleward displacement of the midlatitude storm tracks. The specific study of extreme events over France reveals the high sensitivity of some statistics of climate extremes at the decadal time scale as a consequence of regional climate internal variability.     

Exploring the sample size and replications scenarios effect on spatial prediction of flood,using MARS and MaxEnt methods case study: saliantape catchment,Golestan, Iran

Flood has always been a destructive natural hazard during the recent years. Hence, this research aimed to predict the potentiality and probability of flood phenomenon by using the two well-known models, i.e., the MARS algorithm (multivariate adaptive regression splines) and MaxEnt (maximum entropy) method in the Saliantapeh catchment, Golestan province, Iran, covering 4515.47 km². First, documentary sources report and field surveys were used to provide a flood database map. Then, to prepare the flood spatial potentiality map (FSPM), we select sixteen influential variables as predictors. Furthermore, the relative contributions of predicting factors are estimated using the MaxEnt method. For the analysis of data sensitivity and the uncertainty of the proposed models, different scenarios including the sample size (50%/50%, 80%/20%, and 70%/30%, respectively, for training and validation), and the number of replications (5, 10, and 20) were used. Along with the area under the ROC curve (AUC), the highest accuracy for both models corresponds to the first scenario of sample size (80/20%). Contrarywise, it can be concluded that for this scenario, the MARS technique indicated higher predictive skill (AUC?=?98.51%). Regarding the second scenario, which is corresponding to the replicate, the MARS model with 20 replications still has the highest accuracy (94.70%) compared to the other scenarios and the MaxEnt model. The findings of robustness demonstrated that the scenarios with the greatest AUC value have the highest robustness. This work demonstrates that the utilization of the best accurate model with high certainty along with FSPM may be useful to identify and manage the areas that are most susceptible to flood.

     

Using topographical attributes to evaluate gully erosion proneness (susceptibility) in two mediterranean basins: advantages and limitations

Empirical multivariate predictive models represent an important tool to estimate gully erosion susceptibility. Topography, lithology, climate, land use and vegetation cover are commonly used as input for these approaches. In this paper, two multivariate predictive models were generated for two gully erosion processes in San Giorgio basin (Italy) and Mula River basin (Spain) using only topographical attributes as independent variables. Initially, nine models (five for San Giorgio and four for Mula) with pixel sizes ranging from 2 to 50 m were generated, and validation statistics were calculated to estimate the optimal pixel size. The best models were selected based on model performance using the area under the receiver operating characteristic (AUC) curve and the generalized cross-validation. The best pixel size was 4 m in the San Giorgio basin and 20 m in the Mula basin. The finest resolution was not necessarily the best; rather, the relationship between digital elevation model resolution and size of the landform was important. The two selected models showed an excellent performance with AUC values of 0.859 and 0.826 for San Giorgio and Mula, respectively. The Topographic Wetness Index and the general curvature were identified as key topographical attributes in San Giorgio and Mula basins, respectively. Both attributes were related to the processes observed in the field and described in the literature. Finally, maps of gully erosion susceptibility were produced for each basin. These maps showed that 22 and 20 % of San Giorgio and Mula basins, respectively, present favourable conditions for the development of gullies.     

The role of the diagnostic areas in the assessment of landslide susceptibility models: a test in the sicilian chain

The aim of the research was to verify and compare the predictive power of different diagnostic areas in assessing landslide susceptibility with a multivariate approach. Scarps, landslide areas (the union between scarp and accumulation zones) and areas uphill from crowns, for rotational slides, source or scarp areas and landslide areas, for flows, have been tested. A multivariate approach was applied to assess the landslide susceptibility on the basis of three selected conditioning factors (lithology, slope angle, and topographic wetness index), which were combined in a Unique Condition Unit (UCU) layer. By intersecting the UCU layer with the vector layer of the diagnostic areas, landslide susceptibility models were produced, in which the susceptibility is assigned to each UCUs on the basis of the computed density function. In order to test the effects produced by selecting different diagnostic areas in the performance of the susceptibility models, validation procedures have been applied to evaluate and compare the performances of the derived predictive models. The validation results are estimated by comparing the prediction and the success rate curves, exploiting three morphometric indexes. A test area, the Guddemi river basin, was selected in the northern Sicilian Apennines chain, having a total area of nearly 25 km² and being mainly characterized by the outcropping of clays, calcilutites, and marly limestones. Aerial analysis, integrated with a field survey, resulted in the recognition of 111 earth-flow and 145 earth-rotational slide landslides. Scarps, for rotational slides, and both source and landslide areas, for flows, produced very satisfactory validation results. For rotational slides, areas uphill from crowns and landslide areas are both responsible for lower predictive performances, characterized by validation curves close to being flat shaped, due to their incapability of identifying specific slope (UCU) conditions. Moreover, because of their limited size, the areas uphill from crowns seem to suffer from a relevant geostatistical “instability”, when a splitting is performed to produce the validation domains, so that an enhanced shift between success and prediction rate curves is produced. By comparing the relative susceptibility maps, the research allowed us to evaluate the key role played by the selection of the diagnostic areas; the validation of the models is proposed as a tool to quantify such differences in terms of predictive performance.     

Spatial prediction models for shallow landslide hazards: a comparative assessment of the efficacy of support vector machines,artificial neural networks,kernel logistic regression,and logistic model tree

Preparation of landslide susceptibility maps is considered as the first important step in landslide risk assessments, but these maps are accepted as an end product that can be used for land use planning. The main objective of this study is to explore some new state-of-the-art sophisticated machine learning techniques and introduce a framework for training and validation of shallow landslide susceptibility models by using the latest statistical methods. The Son La hydropower basin (Vietnam) was selected as a case study. First, a landslide inventory map was constructed using the historical landslide locations from two national projects in Vietnam. A total of 12 landslide conditioning factors were then constructed from various data sources. Landslide locations were randomly split into a ratio of 70:30 for training and validating the models. To choose the best subset of conditioning factors, predictive ability of the factors were assessed using the Information Gain Ratio with 10-fold cross-validation technique. Factors with null predictive ability were removed to optimize the models. Subsequently, five landslide models were built using support vector machines (SVM), multi-layer perceptron neural networks (MLP Neural Nets), radial basis function neural networks (RBF Neural Nets), kernel logistic regression (KLR), and logistic model trees (LMT). The resulting models were validated and compared using the receive operating characteristic (ROC), Kappa index, and several statistical evaluation measures. Additionally, Friedman and Wilcoxon signed-rank tests were applied to confirm significant statistical differences among the five machine learning models employed in this study. Overall, the MLP Neural Nets model has the highest prediction capability (90.2 %), followed by the SVM model (88.7 %) and the KLR model (87.9 %), the RBF Neural Nets model (87.1 %), and the LMT model (86.1 %). Results revealed that both the KLR and the LMT models showed promising methods for shallow landslide susceptibility mapping. The result from this study demonstrates the benefit of selecting the optimal machine learning techniques with proper conditioning selection method in shallow landslide susceptibility mapping.     

证据权法在区域滑坡危险性评价中的应用以贵州省为例

以GIS为技术平台,采用证据权法对研究区进行了滑坡地质灾害危险性分析。综合分析历史滑坡数据及其环境因素和触发因素,数据源主要有地形图、DEM、地质图,选取地层岩性、构造、高程、坡度、坡向、地形起伏度、道路、水系作为危险性评价因子。首先应用ArcGIS软件对数据源进行处理,提取各个评价因子图层,并对每个图层进行分级、缓冲区分析等处理,建立若干证据层。然后将历史灾害点与评价因子进行空间关联分析,计算每个评价因子等级的权重,最后计算出评价单元的危险性指数,并将危险性分为极高危险区、高危险区、中等危险区、低危险区。采用成功率曲线法对证据权法评价精度进行验证,结果表明本次评价的精度为71%。利用历史滑坡数据对评价结果进行验证,结果显示评价结果与实际情况较为吻合,说明证据权可以客观定量地评价各影响因子对滑坡的影响程度,该方法应用于区域地质灾害危险性评价比较有效。     

Evaluating hazard results for Switzerland and how not to do it: A discussion of “Problems in the application of the SSHAC probability method for assessing earthquake hazards at Swiss nuclear power plants” by J-U Klügel

The PEGASOS project was a major international seismic hazard study, one of the largest ever conducted anywhere in the world, to assess seismic hazard at four nuclear power plant sites in Switzerland. Before the report of this project has become publicly available, a paper attacking both methodology and results has appeared. Since the general scientific readership may have difficulty in assessing this attack in the absence of the report being attacked, we supply a response in the present paper. The bulk of the attack, besides some misconceived arguments about the role of uncertainties in seismic hazard analysis, is carried by some exercises that purport to be validation exercises. In practice, they are no such thing; they are merely independent sets of hazard calculations based on varying assumptions and procedures, often rather questionable, which come up with various different answers which have no particular significance.     

Support Vector Machines for Landslide Susceptibility Mapping: The Staffora River Basin Case Study, Italy

The aim of this study is the application of support vector machines (SVM) to landslide susceptibility mapping. SVM are a set of machine learning methods in which model capacity matches data complexity. The research is based on a conceptual framework targeted to apply and test all the procedural steps for landslide susceptibility modeling from model selection, to investigation of predictive variables, from empirical cross-validation of results, to analysis of predicted patterns. SVM were successfully applied and the final susceptibility map was interpreted via success and prediction rate curves and receiver operating characteristic (ROC) curves, to support the modeling results and assess the robustness of the model. SVM appeared to be very specific learners, able to discriminate between the informative input and random noise. About 78% of occurrences was identified within the 20% of the most susceptible study area for the cross-validation set. Then the final susceptibility map was compared with other maps, addressed by different statistical approaches, commonly used in susceptibility mapping, such as logistic regression, linear discriminant analysis, and naive Bayes classifier. The SVM procedure was found feasible and able to outperform other techniques in terms of accuracy and generalization capacity. The over-performance of SVM against the other techniques was around 18% for the cross-validation set, considering the 20% of the most susceptible area. Moreover, by analyzing receiver operating characteristic (ROC) curves, SVM appeared to be less prone to false positives than the other models. The study was applied in the Staffora river basin (Lombardy, Northern Italy), an area of about 275 km² characterized by a very high density of landslides, mainly superficial slope failures triggered by intense rainfall events.     

A new GIS-based data mining technique using an adaptive neuro-fuzzy inference system (ANFIS) and k-fold cross-validation approach for land subsidence susceptibility mapping

In this paper, we evaluate the predictive performance of an adaptive neuro-fuzzy inference system (ANFIS) using six different membership functions (MF). In combination with a geographic information system (GIS), ANFIS was used for land subsidence susceptibility mapping (LSSM) in the Marand plain, northwest Iran. This area is prone to droughts and low groundwater levels and subsequent land subsidence damages. Therefore, a land subsidence inventory database was created from an extensive field survey. Areas of land subsidence or areas showing initial signs of subsidence were used for training, while one-third of inventory database were reserved for testing and validation. The inventory database randomly divided into three different folds of the same size. One of the folds was chosen for testing and validation. Other two folds was used for training. This process repeated for every fold in the inventory dataset. Thereafter, land subsidence related factors, such as hydrological and topographical factors, were prepared as GIS layers. Areas susceptible to land subsidence were then analyzed using the ANFIS approach, and land subsidence susceptibility maps were created, whereby six different MFs were applied. Lastly, the results derived from each MF were validated with those areas of the land subsidence database that were not used for training. Receiver operating characteristics (ROC) curves were drawn for all LSSMs, and the areas under the curves were calculated. The ROC analyses for the six LSSMs yielded very high prediction values for two out of the six methods, namely the difference of DsigMF (0.958) and GaussMF (0.951). The integration of ANFIS and GIS generally led to high LSSM prediction accuracies. This study demonstrated that the choice of training dataset and the MF significantly affects the results.