Landslide susceptibility mapping of mountain roads based on machine learning combined model

Landslide susceptibility mapping of mountain roads is frequently confronted by insufficient historical landslide sample data,multicollinearity of existing evaluation index factors,and inconsistency of evaluation factors due to regional environmental variations.Then,a single machine learning model can easily become overfitting,thus reducing the accuracy and robustness of the evaluation model.This paper proposes a combined machine-learning model to address the issues.The landslide susceptibility in mountain roads were mapped by using factor analysis to normalize and reduce the dimensionality of the initial condition factor and generating six new combination factors as evaluation indexes.The mountain roads in the Youxi County,Fujian Province,China were used for the landslide susceptibility mapping.Three most frequently used machine learning techniques,support vector machine(SVM),random forest(RF),and artificial neural network(ANN)models,were used to model the landslide susceptibility of the study area and validate the accuracy of this evaluation index system.The global minimum variance portfolio was utilized to construct a machine learning combined model.5-fold cross-validation,statistical indexes,and AUC(Area Under Curve)values were implemented to evaluate the predictive accuracy of the landslide susceptibility model.The mean AUC values for the SVM,RF,and ANN models in the training stage were 89.2%,88.5%,and 87.9%,respectively,and 78.0%,73.7%,and 76.7%,respectively,in the validating stage.In the training and validation stages,the mean AUC values of the combined model were 92.4% and 87.1%,respectively.The combined model provides greater prediction accuracy and model robustness than one single model.

Mapping landslide susceptibility at the Three Gorges Reservoir, China, using gradient boosting decision tree,random forest and information value models

This work was to generate landslide susceptibility maps for the Three Gorges Reservoir(TGR) area, China by using different machine learning models. Three advanced machine learning methods, namely, gradient boosting decision tree(GBDT), random forest(RF) and information value(InV) models, were used, and the performances were assessed and compared. In total, 202 landslides were mapped by using a series of field surveys, aerial photographs, and reviews of historical and bibliographical data. Nine causative factors were then considered in landslide susceptibility map generation by using the GBDT, RF and InV models. All of the maps of the causative factors were resampled to a resolution of 28.5 m. Of the 486289 pixels in the area,28526 pixels were landslide pixels, and 457763 pixels were non-landslide pixels. Finally, landslide susceptibility maps were generated by using the three machine learning models, and their performances were assessed through receiver operating characteristic(ROC) curves, the sensitivity, specificity,overall accuracy(OA), and kappa coefficient(KAPPA). The results showed that the GBDT, RF and In V models in overall produced reasonable accurate landslide susceptibility maps. Among these three methods, the GBDT method outperforms the other two machine learning methods, which can provide strong technical support for producing landslide susceptibility maps in TGR.

Landslide susceptibility mapping using hybrid random forest with GeoDetector and RFE for factor optimization

The present study aims to develop two hybrid models to optimize the factors and enhance the predictive ability of the landslide susceptibility models.For this,a landslide inventory map was created with 406 historical landslides and 2030 non-landslide points,which was randomly divided into two datasets for model training(70%)and model testing(30%).22 factors were initially selected to establish a landslide factor database.We applied the GeoDetector and recursive feature elimination method(RFE)to address factor optimization to reduce information redundancy and collinearity in the data.Thereafter,the frequency ratio method,multicollinearity test,and interactive detector were used to analyze and evaluate the optimized factors.Subsequently,the random forest(RF)model was used to create a landslide susceptibility map with original and optimized factors.The resultant hybrid models GeoDetector-RF and RFE-RF were evaluated and compared by the area under the receiver operating characteristic curve(AUC)and accuracy.The accuracy of the two hybrid models(0.868 for GeoDetector-RF and 0.869 for RFE-RF)were higher than that of the RF model(0.860),indicating that the hybrid models with factor optimization have high reliability and predictability.Both RFE-RF GeoDetector-RF had higher AUC values,respectively 0.863 and 0.860,than RF(0.853).These results confirm the ability of factor optimization methods to improve the performance of landslide susceptibility models.

Co-seismic Landslide Inventory and Susceptibility Mapping in the 2008 Wenchuan Earthquake Disaster Area,China

The Ms 8.0 May 12,2008 Wenchuan earthquake triggered tens of thousands of landslides.The widespread landslides have caused serious casualties and property losses,and posed a great threat to post-earthquake reconstruction.A spatial database,inventoried 43,842 landslides with a total area of 632 km 2,was developed by interpretation of multi-resolution remote sensing images.The landslides can be classified into three categories:swallow,disrupted slides and falls;deep-seated slides and falls,and rock avalanches.The correlation between landslides distribution and the influencing parameters including distance from co-seismic fault,lithology,slope gradient,elevation,peak ground acceleration(PGA) and distance from drainage were analyzed.The distance from co-seismic fault was the most significant parameter followed by slope gradient and PGA was the least significant one.A logistic regression model combined with bivariate statistical analysis(BSA) was adopted for landslide susceptibility mapping.The study area was classified into five categories of landslide susceptibility:very low,low,medium,high and very high.92.0% of the study area belongs to low and very low categories with corresponding 9.0% of the total inventoried landslides.Medium susceptible zones make up 4.2% of the area with 17.7% of the total landslides.The rest of the area was classified into high and very high categories,which makes up 3.9% of the area with corresponding 73.3% of the total landslides.Although the susceptibility map can reveal the likelihood of future landslides and debris flows,and it is helpful for the rebuilding process and future zoning issues.

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 Une 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 Based Landslide Susceptibility Mapping with Application of Analytical Hierarchy Process in District Ghizer, Gilgit Baltistan Pakistan

District Ghizer is a rugged mountainous territory which experiences several landslides each year. There are 16 major landslide areas and 53 villages that are at high risk to hazards. Keeping in view the severity of natural hazards, the present study was designed to generate landslide susceptibility map based on twelve causative factors viz., slope, aspect, elevation, drainage network, Stream Power Index (SPI), Topographic Wetness Index (TWI), lithological units, fault lines, rainfall, road network, land cover and soil texture. Soil texture was determined by particle size analysis and data for other factors were acquired from freely available sources. Analytical Hierarchy Process (AHP) was employed to identify major landslide causative factors in the district Ghizer. Further, a temporal assessment from 1999 till 2015 was generated to assess the impact of land cover change on landslides. It indicated that the barren soil/ exposed rocks and glaciers have reduced while the vegetation and water classes have shown increment. The total area that lies in moderate to very high landslide susceptible zones was 74.38%, while slope is the main landslide causative factor in the district Ghizer. Validation of the susceptibility map showed 88.1% of the landslides in the study area had occurred in the moderate to very high susceptible zones.

Rapid Susceptibility Mapping of Co-seismic Landslides Triggered by the 2013 Lushan Earthquake Using the Regression Model Developed for the 2008 Wenchuan Earthquake

The primary objective of landslide susceptibility mapping is the prediction of potential landslides in landslide-prone areas.The predictive power of a landslide susceptibility mapping model could be tested in an adjacent area of similar geoenvironmental conditions to find out the reliability.Both the 2008 Wenchuan Earthquake and the 2013 Lushan Earthquake occurred in the Longmen Mountain seismic zone,with similar topographical and geological conditions.The two earthquakes are both featured by thrust fault and similar seismic mechanism.This paper adopted the susceptibility mapping model of co-seismic landslides triggered by Wenchuan earthquake to predict the spatial distribution of landslides induced by Lushan earthquake.Six influencing parameters were taken into consideration: distance from the seismic fault,slope gradient,lithology,distance from drainage,elevation and Peak Ground Acceleration(PGA).The preliminary results suggested that the zones with high susceptibility of coseismic landslides were mainly distributed in the mountainous areas of Lushan,Baoxing and Tianquan counties.The co-seismic landslide susceptibility map was completed in two days after the quake and sent to the field investigators to provide guidance for rescue and relief work.The predictive power of the susceptibility map was validated by ROC curve analysis method using 2037 co-seismic landslides in the epicenter area.The AUC value of 0.710 indicated that the susceptibility model derived from Wenchuan Earthquake landslides showed good accuracy in predicting the landslides triggered by Lushan earthquake.

Landslides susceptibility mapping in Guizhou province based on fuzzy theory

The purpose of this study was to assess the susceptibility of landslides around the area of Guizhou province based on fuzzy theory.In first instance, slope, elevation, lithology, proximity to tectonic lines, proximity to drainage and annual precipitation were taken as independent, causal factors in this study.A landslide hazard evaluation factor system was established by classifying these factors into more subclasses according to some rules.Secondly, a trapezoidal fuzzy number weighting(TFNW) approach was used to assess the importance of six causal factors to landslides in an ArcGIS environment.Thirdly, a landslide susceptibility map was created based on a weighted linear combination model.According to this susceptibility map, the study area was classified into four categories of landslide susceptibility:low, moderate, high and very high.Finally, in order to verify the results obtained, the susceptibility map and the landslide inventory map were combined in the GIS.In addition, the weighting procedure showed that TFNW is an efficient method for weighting causal landslide factors.

Landslide susceptibility mapping using Genetic Algorithm for the Rule Set Production(GARP) model

Investigation on landslide phenomenon is necessary for understanding and delineating the landslide prone and safer places for different land use practices. On this basis, a new model known as genetic algorithm for the rule set production was applied in order to assess its efficacy to obtain a better result and a more precise landslide susceptibility map in Klijanerestagh area of Iran. This study considered twelve landslide conditioning factors(LCF) like altitude, slope, aspect, plan curvature, profile curvature, topographic wetness index(TWI), distance from rivers, faults, and roads, land use/cover, and lithology. For modeling purpose, the Genetic Algorithm for the Rule Set Production(GARP) algorithm was applied in order to produce the landslide susceptibility map. Finally, to evaluate the efficacy of the GARP model, receiver operating characteristics curve as well as the Kappa index were employed. Based on these indices, the GARP model predicted the probability of future landslide incidences with the area under the receiver operating characteristics curve(AUC-ROC) values of 0.932, and 0.907 for training and validating datasets, respectively. In addition, Kappa values for the training and validating datasets were computed as 0.775, and 0.716, respectively. Thus, it can be concluded that the GARP algorithm can be a new but effective method for generating landslide susceptibility maps(LSMs). Furthermore, higher contribution of the lithology, distance from roads, and distance from faults was observed, while lower contribution was attributed to soil, profile curvature, and TWI factors. The introduced methodology in this paper can be suggested for other areas with similar topographical and hydrogeological characteristics for land use planning and reducing the landslide damages.

Geospatial susceptibility mapping of earthquake-induced landslides in Nuweiba area, Gulf of Aqaba, Egypt

Earthquake induced landslides are one of the most severe geo-environmental hazards that cause enormous damage to infrastructure, property, and loss of life in Nuweiba area. This study developed a model for mapping the earthquake-induced landslide susceptibility in Nuweiba area in Egypt with considerations of geological, geomorphological, topographical, and seismological factors. An integrated approach of remote sensing and GIS technologies were applied for that target. Several data sources including Terra SAR-X and SPOT 5 satellite imagery, topographic maps, field data, and other geospatial resources were used to model landslide susceptibility. These data were used specifically to produce important thematic layers contributing to landslide occurrences in the region. A rating scheme was developed to assign ranks for the thematic layers and weights for their classes based on their contribution in landslide susceptibility. The ranks and weights were defined based on the knowledge from field survey and authors experiences related to the study area. The landslide susceptibility map delineates the hazard zones to three relative classes of susceptibility: high, moderate, and low. Therefore, the current approach provides a way to assess landslide hazards and serves for geo-hazard planning and prediction in Nuweiba area.

Application of Chebyshev theorem to data preparation in landslide susceptibility mapping studies:an example from Yenice(Karabük,Turkey)region

Landslide database construction is one of the most crucial stages of the landslide susceptibility mapping studies.Although there are many techniques for preparing landslide database in the literature,representative data selection from huge data sets is a challenging,and,to some extent,a subjective task.Thus,in order to produce reliable landslide susceptibility maps,data-driven,objective and representative database construction is a very important stage for these maps.This study mainly focuses on a landslide database construction task.In this study,it was aimed at building a representative landslide database extraction approach by using Chebyshev theorem to evaluate landslide susceptibility in a landslide prone area in the WesternBlack Sea region of Turkey.The study area was divided into two different parts such as training(Basin 1) and testing areas(Basin 2).A total of nine parameters such as topographical elevation,slope,aspect,planar and profile curvatures,stream power index,distance to drainage,normalized difference vegetation index and topographical wetness index were used in the study.Next,frequency distributions of the considered parameters in both landslide and nonlandslide areas were extracted using different sampling strategies,and a total of nine different landslide databases were obtained.Of these,eight databases were gathered by the methodology proposed by this study based on different standard deviations and algebraic multiplication of raster parameter maps.To evaluate landslide susceptibility,Artificial Neural Network method was used in thestudy area considering the different landslide and nonlandslide data.Finally,to assess the performance of the so-produced landslide susceptibility map based on nine data sets,Area Under Curve(AUC approach was implemented both in Basin 1 and Basin2.The best performances(the greatest AUC values were gathered by the landslide susceptibility map produced by two standard deviation databas extracted by the Chebyshev theorem,as 0.873 and0.761,respectively.Results revealed that th methodology proposed by this study is a powerful and objective approach in landslide susceptibility mapping.

Spatial prediction of landslide susceptibility using GIS-based statistical and machine learning models in Wanzhou County,Three Gorges Reservoir, China

Landslide susceptibility mapping is vital for landslide risk management and urban planning.In this study,we used three statistical models[frequency ratio,certainty factor and index of entropy(IOE)]and a machine learning model[random forest(RF)]for landslide susceptibility mapping in Wanzhou County,China.First,a landslide inventory map was prepared using earlier geotechnical investigation reports,aerial images,and field surveys.Then,the redundant factors were excluded from the initial fourteen landslide causal factors via factor correlation analysis.To determine the most effective causal factors,landslide susceptibility evaluations were performed based on four cases with different combinations of factors("cases").In the analysis,465(70%)landslide locations were randomly selected for model training,and 200(30%)landslide locations were selected for verification.The results showed that case 3 produced the best performance for the statistical models and that case 2 produced the best performance for the RF model.Finally,the receiver operating characteristic(ROC)curve was used to verify the accuracy of each model's results for its respective optimal case.The ROC curve analysis showed that the machine learning model performed better than the other three models,and among the three statistical models,the IOE model with weight coefficients was superior.

Applying deep learning and benchmark machine learning algorithms for landslide susceptibility modelling in Rorachu river basin of Sikkim Himalaya, India

Landslide is considered as one of the most severe threats to human life and property in the hilly areas of the world.The number of landslides and the level of damage across the globe has been increasing over time.Therefore,landslide management is essential to maintain the natural and socio-economic dynamics of the hilly region.Rorachu river basin is one of the most landslide-prone areas of the Sikkim selected for the present study.The prime goal of the study is to prepare landslide susceptibility maps(LSMs)using computer-based advanced machine learning techniques and compare the performance of the models.To properly understand the existing spatial relation with the landslide,twenty factors,including triggering and causative factors,were selected.A deep learning algorithm viz.convolutional neural network model(CNN)and three popular machine learning techniques,i.e.,random forest model(RF),artificial neural network model(ANN),and bagging model,were employed to prepare the LSMs.Two separate datasets including training and validation were designed by randomly taken landslide and nonlandslide points.A ratio of 70:30 was considered for the selection of both training and validation points.Multicollinearity was assessed by tolerance and variance inflation factor,and the role of individual conditioning factors was estimated using information gain ratio.The result reveals that there is no severe multicollinearity among the landslide conditioning factors,and the triggering factor rainfall appeared as the leading cause of the landslide.Based on the final prediction values of each model,LSM was constructed and successfully portioned into five distinct classes,like very low,low,moderate,high,and very high susceptibility.The susceptibility class-wise distribution of landslides shows that more than 90%of the landslide area falls under higher landslide susceptibility grades.The precision of models was examined using the area under the curve(AUC)of the receiver operating characteristics(ROC)curve and statistical methods like root mean square error(RMSE)and mean absolute error(MAE).In both datasets(training and validation),the CNN model achieved the maximum AUC value of 0.903 and 0.939,respectively.The lowest value of RMSE and MAE also reveals the better performance of the CNN model.So,it can be concluded that all the models have performed well,but the CNN model has outperformed the other models in terms of precision.

Landslide inventory and susceptibility assessment using multiple statistical approaches along the Karakoram highway,northern Pakistan

China-Pakistan Economic Corridor(CPEC)is a framework of regional connectivity,which will not only benefit China and Pakistan but will have positive impact on Iran,Afghanistan,India,Central Asian Republic,and the region.The surrounding area in CPEC is prone to frequent disruption by geological hazards mainly landslides in northern Pakistan.Comprehensive landslide inventory and susceptibility assessment are rarely available to utilize for landslide mitigation strategies.This study aims to utilize the high-resolution satellite images to develop a comprehensive landslide inventory and subsequently develop landslide susceptibility maps using multiple techniques.The very high-resolution(VHR)satellite images are utilized to develop a landslide inventory using the visual image classification techniques,historic records and field observations.A total of 1632 landslides are mapped in the area.Four statistical models i.e.,frequency ratio,artificial neural network,weights of evidence and logistic regression were used for landslide susceptibility modeling by comparing the landslide inventory with the topographic parameters,geological features,drainage and road network.The developed landslides susceptibility maps were verified using the area under curve(AUC)method.The prediction power of the model was assessed by the prediction rate curve.The success rate curves show 93%,92.8%,92.7%and 87.4%accuracy of susceptibility maps for frequency ratio,artificial neural network,weights of evidence and logistic regression,respectively.The developed landslide inventory and susceptibility maps can be used for land use planning and landslide mitigation strategies.

Uncertainty pattern in landslide susceptibility prediction modelling:Effects of different landslide boundaries and spatial shape expressions

In some studies on landslide susceptibility mapping(LSM),landslide boundary and spatial shape characteristics have been expressed in the form of points or circles in the landslide inventory instead of the accurate polygon form.Different expressions of landslide boundaries and spatial shapes may lead to substantial differences in the distribution of predicted landslide susceptibility indexes(LSIs);moreover,the presence of irregular landslide boundaries and spatial shapes introduces uncertainties into the LSM.To address this issue by accurately drawing polygonal boundaries based on LSM,the uncertainty patterns of LSM modelling under two different landslide boundaries and spatial shapes,such as landslide points and circles,are compared.Within the research area of Ruijin City in China,a total of 370 landslides with accurate boundary information are obtained,and 10 environmental factors,such as slope and lithology,are selected.Then,correlation analyses between the landslide boundary shapes and selected environmental factors are performed via the frequency ratio(FR)method.Next,a support vector machine(SVM)and random forest(RF)based on landslide points,circles and accurate landslide polygons are constructed as point-,circle-and polygon-based SVM and RF models,respectively,to address LSM.Finally,the prediction capabilities of the above models are compared by computing their statistical accuracy using receiver operating characteristic analysis,and the uncertainties of the predicted LSIs under the above models are discussed.The results show that using polygonal surfaces with a higher reliability and accuracy to express the landslide boundary and spatial shape can provide a markedly improved LSM accuracy,compared to those based on the points and circles.Moreover,a higher degree of uncertainty of LSM modelling is present in the expression of points because there are too few grid units acting as model input variables.Additionally,the expression of the landslide boundary as circles introduces errors in measurement and is not as accurate as the polygonal boundary in most LSM modelling cases.In addition,the results under different conditions show that the polygon-based models have a higher LSM accuracy,with lower mean values and larger standard deviations compared with the point-and circle-based models.Finally,the overall LSM accuracy of the RF is superior to that of the SVM,and similar patterns of landslide boundary and spatial shape affecting the LSM modelling are reflected in the SVM and RF models.

Landslide inventory and susceptibility modelling using geospatial tools,in Hunza-Nagar valley,northern Pakistan

A comprehensive landslide inventory and susceptibility maps are prerequisite for developing and implementing landslide mitigation strategies. Landslide susceptibility maps for the landslides prone regions in northern Pakistan are rarely available. The Hunza-Nagar valley in northern Pakistan is known for its frequent and devastating landslides. In this paper, we have developed a landslide inventory map for Hunza-Nagar valley by using the visual interpretation of the SPOT-5 satellite imagery and mapped a total of 172 landslides. The landslide inventory was subsequently divided into modelling and validation data sets. For the development of landslide susceptibility map seven discrete landslide causative factors were correlated with the landslide inventory map using weight of evidence and frequency ratio statistical models. Four different models of conditional independence were used for the selection of landslide causative factors. The produced landslides susceptibility maps were validated by the success rate and area under curves criteria. The prediction power of the models was also validated with the prediction rate curve. The validation results shows that the success rate curves of the weight of evidence and the frequency models are 82% and 79%, respectively. The prediction accuracy results obtained from this study are 84% for weight of evidence model and 80% for the frequency ratio model. Finally, the landslide susceptibility index maps were classified into five different varying susceptibility zones. The validation and prediction result indicates that the weight of evidence and frequency ratio model are reliable to produce an accurate landslide susceptibility map, which may be helpful for landslides management strategies.

Landslide Susceptibility Assessment of the Youfang Catchment using Logistic Regression

A detailed landslide susceptibility map was produced in the Youfang catchment using logistic regression method with datasets developed for a geographic information system(GIS).Known as one of the most landslide-prone areas in China, the Youfang catchment of Longnan mountain region,which lies in the transitional area among QinghaiTibet Plateau, loess Plateau and Sichuan Basin, was selected as a representative case to evaluate the frequency and distribution of landslides.Statistical relationships for landslide susceptibility assessment were developed using landslide and landslide causative factor databases.Logistic regression(LR)was used to create the landslide susceptibility maps based on a series of available data sources: landslide inventory; distance to drainage systems, faults and roads; slope angle and aspect; topographic elevation and topographical wetness index, and land use.The quality of the landslide susceptibility map produced in this paper was validated and the result can be used fordesigning protective and mitigation measures against landslide hazards.The landslide susceptibility map is expected to provide a fundamental tool for landslide hazards assessment and risk management in the Youfang catchment.

Comparative performance assessment of landslide susceptibility models with presence-only,presenceabsence,and pseudo-absence data

The quality of the data for statistical methods plays an important role in landslide susceptibility mapping.How different data types influence the performance of landslide susceptibility maps is worth studying.The goal of this study was to explore the effects of different data types namely,presence-only(PO),presence-absence(PA),and pseudo-absence(PAs) data,on the predictive capability of landslide susceptibility mapping.This was completed by conducting a case study in the landslide-prone Honghe County in the Yunnan Province of China.A total of 428 landslide PO data points were selected.An equivalent number of nonlandslide locations were generated as PA data by random sampling,and 10,000 sites were uniformly selected at random from each region as PAs data.Three landslide susceptibility models,namely the information value model(IVM),logistic regression(LR) model,and maximum entropy(MaxEnt) model,corresponding to the three data types were investigated.Additionally,the area under the receiver operating characteristic curves(ROC-AUC),seven statistical indices(i.e.accuracy,sensibility,falsepositive rate,specificity,precision,Kappa,and Fmeasure),and a landslide density analysis were used to evaluate model performance regarding landslide susceptibility mapping.Our results indicated that the MaxEnt model using PAs data performed the best and had the highest fitness with the highest ROC-AUC values and statistical indices,followed by the IVM model with only landslide data(PO),and the LR model using PA data.Using PAs data avoided the inherent over-predictive shortcomings of PO data by limiting the predicted area of high-landslide susceptibility.Additionally,the random sampling design of landslide PA data increased the uncertainty of landslide susceptibility mapping and influenced the performance of the model.Therefore,our results suggested that the PAs data sampling provided a useful data type in the absence of high-quality data.Finally,we summarized the principles,advantages,and disadvantages of the three data types to assist with model optimization and the improvement of predicted performance and fitness.

基于不同因子筛选方法的LightGBM-SHAP滑坡易发性研究

以重庆市黔江区为例,选取23个评价因子构建滑坡致灾因子数据库,利用地理探测器与皮尔逊-主成分分析2种因子筛选方法选择因子最优组合;基于Bayesian-LightGBM-SHAP混合模型进行滑坡易发性评价,并对模型精度进行验证,分析影响黔江区滑坡发生的主导因子.初始模型的AUC值为0.801,Pearson Correlation Coefficient-BayesianLightGBM模型AUC值为0.824,GeoDetector-Bayesian-LightGBM模型AUC为0.835;由因子重要性可知,多年平均降雨量、高程、POI核密度与距河流距离是滑坡发生的最主要因子,而输沙指数、水流动力指数与坡位对滑坡的发生影响较弱.因子筛选法-Bayesian-LightGBM相结合的混合模型能够提高模型的准确性,为构建合理因子数据库提供参考框架;通过与因子重要性的结合分析,验证了地理探测器能够准确探测各因子对滑坡发生的贡献值,突出各滑坡地理因子组合之间的相关性,从而探究各因子与滑坡之间的关系.

A robust discretization method of factor screening for landslide susceptibility mapping using convolution neural network,random forest,and logistic regression models

The selection of discretization criteria and interval numbers of landslide-related environmental factors generally fails to quantitatively determine orfilter,resulting in uncertainties and limitations in the performance of machine learning(ML)methods for landslide susceptibility mapping(LSM).The aim of this study is to propose a robust discretization criterion(RDC)to quantify and explore the uncertainty and subjectivity of different discretization methods.The RDC consists of two steps:raw classification dataset generation and optimal dataset extraction.To evaluate the robustness of the proposed RDC method,Lushan County of Sichuan Province in China was chosen as the study area to generate the LSM based on three datasets(optimal dataset,original dataset with continuous values,and statistical dataset)using three popular ML methods,namely,convolution neural network,random forest,and logistic regression.The results show that the areas under the receiver operating characteristic curve(AUCs)of the optimal dataset for the abovementioned ML models are 0.963,0.961,and 0.930 which are higher than those of the original dataset(0.938,0.947,and 0.900)and statistical dataset(0.948,0.954,and 0.897).In conclusion,the RDC method can extract the more representative features from environmental factors and outperform the other conventional discretization methods.