Machine learning is currently one of the research hotspots in the field of landslide prediction.To clarify and evaluate the differences in characteristics and prediction effects of different machine learning models,Co...Machine learning is currently one of the research hotspots in the field of landslide prediction.To clarify and evaluate the differences in characteristics and prediction effects of different machine learning models,Conghua District,which is the most prone to landslide disasters in Guangzhou,was selected for landslide susceptibility evaluation.The evaluation factors were selected by using correlation analysis and variance expansion factor method.Applying four machine learning methods namely Logistic Regression(LR),Random Forest(RF),Support Vector Machines(SVM),and Extreme Gradient Boosting(XGB),landslide models were constructed.Comparative analysis and evaluation of the model were conducted through statistical indices and receiver operating characteristic(ROC)curves.The results showed that LR,RF,SVM,and XGB models have good predictive performance for landslide susceptibility,with the area under curve(AUC)values of 0.752,0.965,0.996,and 0.998,respectively.XGB model had the highest predictive ability,followed by RF model,SVM model,and LR model.The frequency ratio(FR)accuracy of LR,RF,SVM,and XGB models was 0.775,0.842,0.759,and 0.822,respectively.RF and XGB models were superior to LR and SVM models,indicating that the integrated algorithm has better predictive ability than a single classification algorithm in regional landslide classification problems.展开更多
The applicability of statistics-based landslide susceptibility assessment methods is affected by the number of historical landslides.Previous studies have proposed support vector machine(SVM)as a small-sample learning...The applicability of statistics-based landslide susceptibility assessment methods is affected by the number of historical landslides.Previous studies have proposed support vector machine(SVM)as a small-sample learning method.However,those studies demonstrated that different parameters can affect model performance.We optimized the SVM and obtained models as 5-fold cross validation(5-CV)SVM,genetic algorithm(GA)SVM,and particle swarm optimization(PSO)SVM.This study compared the prediction performances of logistic regression(LR),5-CV SVM,GA SVM,and PSO SVM on landslide susceptibility mapping,to explore the spatial distribution of landslide susceptibility in the study area in Tibetan Plateau,China.A geospatial database was established based on 392 historical landslides and 392 non-landslides in the study area.We used 11 influencing factors of altitude,slope,aspect,curvature,lithology,normalized difference vegetation index(NDVI),distance to road,distance to river,distance to fault,peak ground acceleration(PGA),and rainfall to construct an influencing factor evaluation system.To evaluate the models,four susceptibility maps were compared via receiver operating characteristics(ROC)curve and the results showed that prediction rates for the models are 84%(LR),87%(5-CV SVM),85%(GA SVM),and 90%(PSO SVM).We also used precision,recall,F1-score and accuracy to assess the quality performance of these models.The results showed that the PSO SVM had greater potential for future implementation in the Tibetan Plateau area because of its superior performance in the landslide susceptibility assessment.展开更多
Bivariate statistical analysis of data-driven approaches is widely used for landslide susceptibility assessment, and the frequency ratio(FR) method is one of the most popular. However, the results of such assessments ...Bivariate statistical analysis of data-driven approaches is widely used for landslide susceptibility assessment, and the frequency ratio(FR) method is one of the most popular. However, the results of such assessments are dominated by the number of classes and bounds of landslide-related causative factors, and the optimal assessment is unknown. This paper optimizes the frequency ratio method as an example of bivariate statistical analysis for landslide susceptibility mapping based on a case study of the Caiyuan Basin, a region with frequent landslides, which is located in the southeast coastal mountainous area of China. A landslide inventory map containing a total of 1425 landslides(polygons) was produced, in which 70% of the landslides were selected for training purposes, and the remaining were used for validationpurposes. All datasets were resampled to the same 5 m × 5 m/pixel resolution. The receiver operating characteristic(ROC) curves of the susceptibility maps were obtained based on different combinations of dominating parameters, and the maximum value of the areas under the ROC curves(AUCs) as well as the corresponding optimal parameter was identified with an automatic searching algorithm. The results showed that the landslide susceptibility maps obtained using optimal parameters displayed a significant increase in the prediction AUC compared with those values obtained using stochastic parameters. The results also showed that one parameter named bin width has a dominant influence on the optimum. In practice, this paper is expected to benefit the assessment of landslide susceptibility by providing an easy-to-use tool. The proposed automatic approach provides a way to optimize the frequency ratio method or other bivariate statistical methods, which can furtherfacilitate comparisons and choices between different methods for landslide susceptibility assessment.展开更多
This study presents a statistical landslide susceptibility assessment(LSA) in a dynamic environment. The study area is located in the eastern part of Lanzhou, NW China. The Lanzhou area has exhibited rapid urbanizatio...This study presents a statistical landslide susceptibility assessment(LSA) in a dynamic environment. The study area is located in the eastern part of Lanzhou, NW China. The Lanzhou area has exhibited rapid urbanization rates over the past decade associated with greening, continuous land use change, and geomorphic reshaping activities. To consider the dynamics of the environment in the LSA, multitemporal data for landslide inventories and the corresponding causal factors were collected. The weights of evidence(Wof E) method was used to perform the LSA. Three time stamps, i.e., 2000, 2012, and 2016, were selected to assess the state of landslide susceptibility over time. The results show a clear evolution of the landslide susceptibility patterns that was mainly governed by anthropogenic activities directed toward generating safer building grounds for civil infrastructure. The low and very low susceptibility areas increased by approximately 10% between 2000 and 2016. At the same time, areas of medium, high and very high susceptibility zones decreased proportionally. Based on the results, an approach to design the statistical LSA under dynamic conditions is proposed, the issues and limitations of this approach are also discussed. The study shows that under dynamic conditions, the requirements for data quantity and quality increase significantly. A dynamic environment requires greater effort to estimate the causal relations between the landslides and controlling factors as well as for model validation.展开更多
Reliable assessment of landslide susceptibility in broad areas of terrain remains challenging due to complex topography and poor representation of randomly selected negative samples.Assessment in broad areas is now pr...Reliable assessment of landslide susceptibility in broad areas of terrain remains challenging due to complex topography and poor representation of randomly selected negative samples.Assessment in broad areas is now primarily based on grid units,which do not have a clear physical meaning like slope units,and their accuracy is not ideal.Nevertheless,the large amount of manual editing,due to the incorrectly generated horizontal and vertical lines during slope unit partitioning,limits using slope units for rapid assessment over large areas.Hence,this paper proposes a reliable susceptibility assessment approach to solve this problem based on optimal slope units and negative samples involving prior knowledge.Precisely,an algorithm to automatically extract slope units is designed to eliminate fragmented and erroneous units.Second,a samples labeling index(SLI)is defined based on the certainty factors model to select negative samples reasonably.Sichuan Province,China is selected for experimental analysis,with the results demonstrate that the optimized slope unit and the negative samples selection strategy consider prior knowledge achieve better results in the random forest model,support vector machine model,and artificial neural network model.In particular,the composite performance index AUC of artificial neural network model improved from 0.81 to 0.90.展开更多
Landslides are widespread geomorphological phenomena with complex mechanisms that have caused extensive causalities and property damage worldwide.The scale and frequency of landslides are presently increasing owing to...Landslides are widespread geomorphological phenomena with complex mechanisms that have caused extensive causalities and property damage worldwide.The scale and frequency of landslides are presently increasing owing to the warming effects of climate change,which further increases the associated safety risks.In this study,the relationship between historical landslides and environmental variables in the Hanjiang River Basin was determined and an optimized model was used to constrain the relative contribution of variables and best spatial response curve.The optimal MaxEnt model was used to predict the current distribution of landslides and influence of future rainfall changes on the landslide susceptibility.The results indicate that environmental variables in the study area statistically correlate with landslide events over the past 20 years.The MaxEnt model evaluation was applied to landslide hazards in the Hanjiang River Basin based on current climate change scenarios.The results indicate that 25.9%of the study area is classified as a high-risk area.The main environmental variables that affect the distribution of landslides include altitude,slope,normalized difference vegetation index,annual precipitation,distance from rivers,and distance from roads,with a cumulative contribution rate of approximately 90%.The annual rainfall in the Hanjiang River Basin will continue to increase under future climate warming scenarios.Increased rainfall will further increase the extent of high-and medium-risk areas in the basin,especially when following the RCP8.5 climate prediction,which is expected to increase the high-risk area by 10.7%by 2070.Furthermore,high landslide risk areas in the basin will migrate to high-altitude areas in the future,which poses new challenges for the prevention and control of landslide risks.This study demonstrates the usefulness of the MaxEnt model as a tool for landslide susceptibility prediction in the Hanjiang River Basin caused by global warming and yields robust prediction results.This approach therefore provides an important reference for river basin management and disaster reduction and prevention.The study on landslide risks also supports the hypothesis that global climate change will further enhance the frequency and intensity of landslide activity throughout the course of the 21st Century.展开更多
基金supported by the projects of the China Geological Survey(DD20221729,DD20190291)Zhuhai Urban Geological Survey(including informatization)(MZCD–2201–008).
文摘Machine learning is currently one of the research hotspots in the field of landslide prediction.To clarify and evaluate the differences in characteristics and prediction effects of different machine learning models,Conghua District,which is the most prone to landslide disasters in Guangzhou,was selected for landslide susceptibility evaluation.The evaluation factors were selected by using correlation analysis and variance expansion factor method.Applying four machine learning methods namely Logistic Regression(LR),Random Forest(RF),Support Vector Machines(SVM),and Extreme Gradient Boosting(XGB),landslide models were constructed.Comparative analysis and evaluation of the model were conducted through statistical indices and receiver operating characteristic(ROC)curves.The results showed that LR,RF,SVM,and XGB models have good predictive performance for landslide susceptibility,with the area under curve(AUC)values of 0.752,0.965,0.996,and 0.998,respectively.XGB model had the highest predictive ability,followed by RF model,SVM model,and LR model.The frequency ratio(FR)accuracy of LR,RF,SVM,and XGB models was 0.775,0.842,0.759,and 0.822,respectively.RF and XGB models were superior to LR and SVM models,indicating that the integrated algorithm has better predictive ability than a single classification algorithm in regional landslide classification problems.
基金financially supported by the National Natural Science Foundation of China(41977213)the Second Tibetan Plateau Scientific Expedition and Research Program(2019QZKK0906)+3 种基金Science and Technology Department of Sichuan Province(2021YJ0032)Sichuan Transportation Science and Technology Project(2021-A-03)Sichuan Science and Technology Program(2022NSFSC0425)CREC Sichuan Eco-City Investment Co,Ltd.(R110121H01092)。
文摘The applicability of statistics-based landslide susceptibility assessment methods is affected by the number of historical landslides.Previous studies have proposed support vector machine(SVM)as a small-sample learning method.However,those studies demonstrated that different parameters can affect model performance.We optimized the SVM and obtained models as 5-fold cross validation(5-CV)SVM,genetic algorithm(GA)SVM,and particle swarm optimization(PSO)SVM.This study compared the prediction performances of logistic regression(LR),5-CV SVM,GA SVM,and PSO SVM on landslide susceptibility mapping,to explore the spatial distribution of landslide susceptibility in the study area in Tibetan Plateau,China.A geospatial database was established based on 392 historical landslides and 392 non-landslides in the study area.We used 11 influencing factors of altitude,slope,aspect,curvature,lithology,normalized difference vegetation index(NDVI),distance to road,distance to river,distance to fault,peak ground acceleration(PGA),and rainfall to construct an influencing factor evaluation system.To evaluate the models,four susceptibility maps were compared via receiver operating characteristics(ROC)curve and the results showed that prediction rates for the models are 84%(LR),87%(5-CV SVM),85%(GA SVM),and 90%(PSO SVM).We also used precision,recall,F1-score and accuracy to assess the quality performance of these models.The results showed that the PSO SVM had greater potential for future implementation in the Tibetan Plateau area because of its superior performance in the landslide susceptibility assessment.
基金funded by the National Natural Science Foundation of China(Grant NO.41525010,41807291,41421001,41790443 and 41701458)the Strategic Priority Research Program of Chinese Academy of Sciences(CAS)(Grant NO.XDA23090301 and XDA19040304)+1 种基金the Key Research Program of Frontier Sciences of Chinese Academy of Sciences(CAS)(Grant NO.QYZDY-SSW-DQC019)the Second Tibetan Plateau Scientific Expedition and Research(STEP)program(Grant No.2019QZKK0904)
文摘Bivariate statistical analysis of data-driven approaches is widely used for landslide susceptibility assessment, and the frequency ratio(FR) method is one of the most popular. However, the results of such assessments are dominated by the number of classes and bounds of landslide-related causative factors, and the optimal assessment is unknown. This paper optimizes the frequency ratio method as an example of bivariate statistical analysis for landslide susceptibility mapping based on a case study of the Caiyuan Basin, a region with frequent landslides, which is located in the southeast coastal mountainous area of China. A landslide inventory map containing a total of 1425 landslides(polygons) was produced, in which 70% of the landslides were selected for training purposes, and the remaining were used for validationpurposes. All datasets were resampled to the same 5 m × 5 m/pixel resolution. The receiver operating characteristic(ROC) curves of the susceptibility maps were obtained based on different combinations of dominating parameters, and the maximum value of the areas under the ROC curves(AUCs) as well as the corresponding optimal parameter was identified with an automatic searching algorithm. The results showed that the landslide susceptibility maps obtained using optimal parameters displayed a significant increase in the prediction AUC compared with those values obtained using stochastic parameters. The results also showed that one parameter named bin width has a dominant influence on the optimum. In practice, this paper is expected to benefit the assessment of landslide susceptibility by providing an easy-to-use tool. The proposed automatic approach provides a way to optimize the frequency ratio method or other bivariate statistical methods, which can furtherfacilitate comparisons and choices between different methods for landslide susceptibility assessment.
基金the framework of a scientific-technical cooperation project between the Federal Institute for Geosciences and Natural Resources(BGR)and the China Geological Survey(CGS)co-funded by the German Ministry of the Economic Affairs and Energy(BMWi)and Ministry of Land and Resources of the People's Republik of China
文摘This study presents a statistical landslide susceptibility assessment(LSA) in a dynamic environment. The study area is located in the eastern part of Lanzhou, NW China. The Lanzhou area has exhibited rapid urbanization rates over the past decade associated with greening, continuous land use change, and geomorphic reshaping activities. To consider the dynamics of the environment in the LSA, multitemporal data for landslide inventories and the corresponding causal factors were collected. The weights of evidence(Wof E) method was used to perform the LSA. Three time stamps, i.e., 2000, 2012, and 2016, were selected to assess the state of landslide susceptibility over time. The results show a clear evolution of the landslide susceptibility patterns that was mainly governed by anthropogenic activities directed toward generating safer building grounds for civil infrastructure. The low and very low susceptibility areas increased by approximately 10% between 2000 and 2016. At the same time, areas of medium, high and very high susceptibility zones decreased proportionally. Based on the results, an approach to design the statistical LSA under dynamic conditions is proposed, the issues and limitations of this approach are also discussed. The study shows that under dynamic conditions, the requirements for data quantity and quality increase significantly. A dynamic environment requires greater effort to estimate the causal relations between the landslides and controlling factors as well as for model validation.
基金supported by the National Natural Science Foundation of China[grant number 41941019]Identification of potential geohazards by integrated remote sensing technologies and applications[grant number DD20211365].
文摘Reliable assessment of landslide susceptibility in broad areas of terrain remains challenging due to complex topography and poor representation of randomly selected negative samples.Assessment in broad areas is now primarily based on grid units,which do not have a clear physical meaning like slope units,and their accuracy is not ideal.Nevertheless,the large amount of manual editing,due to the incorrectly generated horizontal and vertical lines during slope unit partitioning,limits using slope units for rapid assessment over large areas.Hence,this paper proposes a reliable susceptibility assessment approach to solve this problem based on optimal slope units and negative samples involving prior knowledge.Precisely,an algorithm to automatically extract slope units is designed to eliminate fragmented and erroneous units.Second,a samples labeling index(SLI)is defined based on the certainty factors model to select negative samples reasonably.Sichuan Province,China is selected for experimental analysis,with the results demonstrate that the optimized slope unit and the negative samples selection strategy consider prior knowledge achieve better results in the random forest model,support vector machine model,and artificial neural network model.In particular,the composite performance index AUC of artificial neural network model improved from 0.81 to 0.90.
基金funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)National Foundation of Forestry Science and Technology Popularization(No.[2015]17)Major Fund for Natural Science of Jiangsu Higher Education Institutions(No.15KJA220004).
文摘Landslides are widespread geomorphological phenomena with complex mechanisms that have caused extensive causalities and property damage worldwide.The scale and frequency of landslides are presently increasing owing to the warming effects of climate change,which further increases the associated safety risks.In this study,the relationship between historical landslides and environmental variables in the Hanjiang River Basin was determined and an optimized model was used to constrain the relative contribution of variables and best spatial response curve.The optimal MaxEnt model was used to predict the current distribution of landslides and influence of future rainfall changes on the landslide susceptibility.The results indicate that environmental variables in the study area statistically correlate with landslide events over the past 20 years.The MaxEnt model evaluation was applied to landslide hazards in the Hanjiang River Basin based on current climate change scenarios.The results indicate that 25.9%of the study area is classified as a high-risk area.The main environmental variables that affect the distribution of landslides include altitude,slope,normalized difference vegetation index,annual precipitation,distance from rivers,and distance from roads,with a cumulative contribution rate of approximately 90%.The annual rainfall in the Hanjiang River Basin will continue to increase under future climate warming scenarios.Increased rainfall will further increase the extent of high-and medium-risk areas in the basin,especially when following the RCP8.5 climate prediction,which is expected to increase the high-risk area by 10.7%by 2070.Furthermore,high landslide risk areas in the basin will migrate to high-altitude areas in the future,which poses new challenges for the prevention and control of landslide risks.This study demonstrates the usefulness of the MaxEnt model as a tool for landslide susceptibility prediction in the Hanjiang River Basin caused by global warming and yields robust prediction results.This approach therefore provides an important reference for river basin management and disaster reduction and prevention.The study on landslide risks also supports the hypothesis that global climate change will further enhance the frequency and intensity of landslide activity throughout the course of the 21st Century.
