scholarly journals Exploring the potential relationship between the occurrence of debris flow and landslide

2020 ◽  
Author(s):  
Zhu Liang ◽  
Changming Wang ◽  
Donghe Ma ◽  
Kaleem Ullah Jan Khan
Keyword(s):  
Random Forest ◽  
Debris Flow ◽  
Characteristic Curve ◽  
Area Under The Curve ◽  
Susceptibility Map ◽  
Relative Operating Characteristic ◽  
Conditioning Factors ◽  
Statistical Measures ◽  
Potential Relationship ◽  
Major Factors

Abstract. he aim of the present study is to explore the potential relationship between debris flow and soil slide by establishing susceptibility zoning maps (SZM) separately with the use of random forest. Longzi County, located in Southeastern Tibet, where historical landslides occurred commonly, was selected as the study area. The work has been carried out with the following steps: (1) An inventory map consisting of 448 landslides (399 soil slides and 49 debris flows) was determined; (2) Slope units and 11 conditioning factors were prepared for the susceptibility modelling of landslide while watershed units and 12 factors for debris flow; (3) SZM were constructed for landslide and debris flow, respectively, with the use of random forest; (4) The performance of two models were evaluated by 5-fold cross-validation using relative operating characteristic curve (ROC), area under the curve (AUC) and statistical measures; (5) The potential relationship between soil slide and debris flow was explored by the superimposition of two zoning maps; (6) Gini index was applied to determined the major factors and analyze the difference between debris flow and soil slide; (7) A combined susceptibility map with two kinds of disaster was obtained. Two models had demonstrated great predictive capabilities, of which accuracy and AUC was 87.33 %, 0.902 and 85.17 %, 0.892, respectively. The loose sources need by the debris flow were not necessarily brought by the landslides although most landslides can be converted into debris flow. The area prone to debris flow did not promote the occurrence of landslide. A susceptibility zoning map composed of two or more natural disasters is comprehensive and significant in this regard, which provides valuable reference for researches of disaster-chain and engineering applications.

scholarly journals Exploring the potential relationship between the occurrence of debris flow and landslides

2021 ◽  
Vol 21 (4) ◽  
pp. 1247-1262
Author(s):  
Zhu Liang ◽  
Changming Wang ◽  
Donghe Ma ◽  
Kaleem Ullah Jan Khan
Keyword(s):  
Debris Flow ◽  
Area Under The Curve ◽  
Susceptibility Map ◽  
Conditioning Factors ◽  
Susceptibility Modeling ◽  
Statistical Measures ◽  
Potential Relationship ◽  
The Difference ◽  
Roc Area ◽  
Major Factors

Abstract. The present study is to explore the potential relationship between debris flow and landslides by establishing susceptibility zoning maps (SZMs) separately with the use of random forest (RF). Lhünzê county, located in southeastern Tibet, was selected as the study area. The work was carried out with the following steps: (1) an inventory map consisting of 399 landslides and 49 debris flows was determined; (2) slope units and 11 conditioning factors were prepared for the susceptibility modeling of landslide while watershed units and 12 factors were prepared for debris flow; (3) SZMs were constructed for landslide and debris flow, respectively, with the use of RF; (4) the performance of two models was evaluated by 5-fold cross-validation using receiver operating characteristic (ROC), area under the curve (AUC) and statistical measures; (5) the potential relationship between landslide and debris flow was explored by the superimposition of two zoning maps; (6) the Gini index was applied to determine the major factors and analyze the difference between debris flow and landslides; (7) a combined susceptibility map with two considered hazardous phenomena was obtained. Two used models had demonstrated great predictive capabilities, with an accuracy of 87.33 % and 85.17 % and AUC of 0.902 and 0.892, respectively. Comparing the overlap of different susceptibility classes for two obtained maps, it was concluded that there is no straightforward relationship between the occurrence of debris flow and landslides. Although most landslides can be converted into debris flow, the area prone to debris flow did not promote the occurrence of a landslide. A susceptibility zoning map composed of two or more hazardous phenomena is comprehensive and significant in this regard, which provides a valuable reference for research studies of disaster-chain and engineering applications.

scholarly journals Postprocessing Next-Day Ensemble Probabilistic Precipitation Forecasts Using Random Forests

2019 ◽  
Vol 34 (6) ◽  
pp. 2017-2044 ◽  
Author(s):  
Eric D. Loken ◽  
Adam J. Clark ◽  
Amy McGovern ◽  
Montgomery Flora ◽  
Kent Knopfmeier
Keyword(s):  
Random Forest ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
Ensemble Forecast ◽  
Grid Spacing ◽  
Relative Operating Characteristic ◽  
Advantages And Disadvantages ◽  
Skill Scores ◽  
Systematic Biases ◽  
Fold Cross Validation

Abstract Most ensembles suffer from underdispersion and systematic biases. One way to correct for these shortcomings is via machine learning (ML), which is advantageous due to its ability to identify and correct nonlinear biases. This study uses a single random forest (RF) to calibrate next-day (i.e., 12–36-h lead time) probabilistic precipitation forecasts over the contiguous United States (CONUS) from the Short-Range Ensemble Forecast System (SREF) with 16-km grid spacing and the High-Resolution Ensemble Forecast version 2 (HREFv2) with 3-km grid spacing. Random forest forecast probabilities (RFFPs) from each ensemble are compared against raw ensemble probabilities over 496 days from April 2017 to November 2018 using 16-fold cross validation. RFFPs are also compared against spatially smoothed ensemble probabilities since the raw SREF and HREFv2 probabilities are overconfident and undersample the true forecast probability density function. Probabilistic precipitation forecasts are evaluated at four precipitation thresholds ranging from 0.1 to 3 in. In general, RFFPs are found to have better forecast reliability and resolution, fewer spatial biases, and significantly greater Brier skill scores and areas under the relative operating characteristic curve compared to corresponding raw and spatially smoothed ensemble probabilities. The RFFPs perform best at the lower thresholds, which have a greater observed climatological frequency. Additionally, the RF-based postprocessing technique benefits the SREF more than the HREFv2, likely because the raw SREF forecasts contain more systematic biases than those from the raw HREFv2. It is concluded that the RFFPs provide a convenient, skillful summary of calibrated ensemble output and are computationally feasible to implement in real time. Advantages and disadvantages of ML-based postprocessing techniques are discussed.

scholarly journals Shallow Landslide Susceptibility Mapping by Random Forest Base Classifier and Its Ensembles in a Semi-Arid Region of Iran

Forests ◽  
2020 ◽  
Vol 11 (4) ◽  
pp. 421 ◽  
Author(s):  
Viet-Ha Nhu ◽  
Ataollah Shirzadi ◽  
Himan Shahabi ◽  
Wei Chen ◽  
John J Clague ◽  
...  
Keyword(s):  
Random Forest ◽  
Landslide Susceptibility ◽  
Goodness Of Fit ◽  
Information Gain ◽  
Characteristic Curve ◽  
Model Performance ◽  
Shallow Landslide ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Conditioning Factors

We generated high-quality shallow landslide susceptibility maps for Bijar County, Kurdistan Province, Iran, using Random Forest (RAF), an ensemble computational intelligence method and three meta classifiers—Bagging (BA, BA-RAF), Random Subspace (RS, RS-RAF), and Rotation Forest (RF, RF-RAF). Modeling and validation were done on 111 shallow landslide locations using 20 conditioning factors tested by the Information Gain Ratio (IGR) technique. We assessed model performance with statistically based indexes, including sensitivity, specificity, accuracy, kappa, root mean square error (RMSE), and area under the receiver operatic characteristic curve (AUC). All four machine learning models that we tested yielded excellent goodness-of-fit and prediction accuracy, but the RF-RAF ensemble model (AUC = 0.936) outperformed the BA-RAF, RS-RAF (AUC = 0.907), and RAF (AUC = 0.812) models. The results also show that the Random Forest model significantly improved the predictive capability of the RAF-based classifier and, therefore, can be considered as a useful and an effective tool in regional shallow landslide susceptibility mapping.

scholarly journals Random forest–based feature selection and detection method for drunk driving recognition

2020 ◽  
Vol 16 (2) ◽  
pp. 155014772090523
Author(s):  
ZhenLong Li ◽  
HaoXin Wang ◽  
YaoWei Zhang ◽  
XiaoHua Zhao
Keyword(s):  
Feature Selection ◽  
Random Forest ◽  
Driving Simulator ◽  
Characteristic Curve ◽  
Area Under The Curve ◽  
Drunk Driving ◽  
Support Vector ◽  
Linear Discriminant ◽  
Dummy Variable ◽  
University Of Technology

A method for drunk driving detection using Feature Selection based on the Random Forest was proposed. First, driving behavior data were collected using a driving simulator at Beijing University of Technology. Second, the features were selected according to the Feature Importance in the random forest. Third, a dummy variable was introduced to encode the geometric characteristics of different roads so that drunk driving under different road conditions can be detected with the same classifier based on the random forest. Finally, the linear discriminant analysis, support vector machine, and AdaBoost classifiers were used and compared with the random forest. The accuracy, F1 score, receiver operating characteristic curve, and area under the curve value were used to evaluate the performance of the classifiers. The results show that Accelerator Depth, Speed, Distance to the Center of the Lane, Acceleration, Engine Revolution, Brake Depth, and Steering Angle have important influences on identifying the drivers’ states and can be used to detect drunk driving. Specifically, the classifiers with Accelerator Depth outperformed the other classifiers without Accelerator Depth. This means that Accelerator Depth is an important feature. Both the AdaBoost and random forest classifiers have an accuracy of 81.48%, which verified the effectiveness of the proposed method.

scholarly journals Sinkhole Detection and Characterization Using LiDAR-Derived DEM with Logistic Regression

2019 ◽  
Vol 11 (13) ◽  
pp. 1592 ◽  
Author(s):  
Yong Je Kim ◽  
Boo Hyun Nam ◽  
Heejung Youn
Keyword(s):  
Logistic Regression ◽  
Characteristic Curve ◽  
Area Under The Curve ◽  
Susceptibility Map ◽  
Cutoff Value ◽  
Geometric Characteristics ◽  
Probabilistic Detection ◽  
Remote Sensing Techniques ◽  
Karst Landscapes ◽  
Elevation Model

Depressions due to sinkhole formation cause significant structural damages to buildings and civil infrastructure. Traditionally, visual inspection has been used to detect sinkholes, which is a subjective way and time- and labor-consuming. Remote sensing techniques have been introduced for morphometric studies of karst landscapes. This study presents a methodology for the probabilistic detection of sinkholes using LiDAR-derived digital elevation model (DEM) data. The proposed study provides benefits associated with: (1) Detection of unreported sinkholes in rural and/or inaccessible areas, (2) automatic delineation of sinkhole boundaries, and (3) quantification of the geometric characteristics of those identified sinkholes. Among sixteen morphometric parameters, nine parameters were chosen for logistic regression, which was then employed to compute the probability of sinkhole detection; a cutoff value was back-calculated such that the sinkhole susceptibility map well predicted the reported sinkhole boundaries. According to the results of the LR model, the optimal cutoff value was calculated to be 0.13, and the area under the curve (AUC) of the receiver operating characteristic curve (ROC) was 0.90, indicating the model is reliable for the study area. For those identified sinkholes, the geometric characteristics (e.g., depth, length, area, and volume) were computed.

scholarly journals New Ensemble Models for Shallow Landslide Susceptibility Modeling in a Semi-Arid Watershed

Forests ◽  
2019 ◽  
Vol 10 (9) ◽  
pp. 743 ◽  
Author(s):  
Dieu Tien Bui ◽  
Ataollah Shirzadi ◽  
Himan Shahabi ◽  
Marten Geertsema ◽  
Ebrahim Omidvar ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Landslide Susceptibility ◽  
Goodness Of Fit ◽  
Model Comparison ◽  
Characteristic Curve ◽  
Support Vector ◽  
Susceptibility Map ◽  
Random Subspace ◽  
Ensemble Models ◽  
Conditioning Factors

We prepared a landslide susceptibility map for the Sarkhoon watershed, Chaharmahal-w-bakhtiari, Iran, using novel ensemble artificial intelligence approaches. A classifier of support vector machine (SVM) was employed as a base classifier, and four Meta/ensemble classifiers, including Adaboost (AB), bagging (BA), rotation forest (RF), and random subspace (RS), were used to construct new ensemble models. SVM has been used previously to spatially predict landslides, but not together with its ensembles. We selected 20 conditioning factors and randomly portioned 98 landslide locations into training (70%) and validating (30%) groups. Several statistical metrics, including sensitivity, specificity, accuracy, kappa, root mean square error (RMSE), and area under the receiver operatic characteristic curve (AUC), were used for model comparison and validation. Using the One-R Attribute Evaluation (ORAE) technique, we found that all 20 conditioning factors were significant in identifying landslide locations, but “distance to road” was found to be the most important. The RS (AUC = 0.837) and RF (AUC = 0.834) significantly improved the goodness-of-fit and prediction accuracy of the SVM (AUC = 0.810), whereas the BA (AUC = 0.807) and AB (AUC = 0.779) did not. The random subspace based support vector machine (RSSVM) model is a promising technique for helping to better manage land in landslide-prone areas.

scholarly journals A Comparative Assessment of Random Forest and k-Nearest Neighbor Classifiers for Gully Erosion Susceptibility Mapping

Water ◽  
2019 ◽  
Vol 11 (10) ◽  
pp. 2076 ◽  
Author(s):  
Avand ◽  
Janizadeh ◽  
Naghibi ◽  
Pourghasemi ◽  
Khosrobeigi Bozchaloei ◽  
...  
Keyword(s):  
Random Forest ◽  
Nearest Neighbor ◽  
Area Under The Curve ◽  
Gully Erosion ◽  
K Nearest Neighbor ◽  
K Nearest Neighbors ◽  
Factors Affecting ◽  
Conditioning Factors ◽  
Susceptibility Maps ◽  
Nearest Neighbor Classifiers

This research was conducted to determine which areas in the Robat Turk watershed in Iran are sensitive to gully erosion, and to define the relationship between gully erosion and geo-environmental factors by two data mining techniques, namely, Random Forest (RF) and k-Nearest Neighbors (KNN). First, 242 gully locations we determined in field surveys and mapped in ArcGIS software. Then, twelve gully-related conditioning factors were selected. Our results showed that, for both the RF and KNN models, altitude, distance to roads, and distance from the river had the highest influence upon gully erosion sensitivity. We assessed the gully erosion susceptibility maps using the Receiver Operating Characteristic (ROC) curve. Validation results showed that the RF and KNN models had Area Under the Curve (AUC) 87.4 and 80.9%, respectively. As a result, the RF method has better performance compared with the KNN method for mapping gully erosion susceptibility. Rainfall, altitude, and distance from a river were identified as the most important factors affecting gully erosion in this area. The methodology used in this research is transferable to other regions to determine which areas are prone to gully erosion and to explicitly delineate high-risk zones within these areas.

scholarly journals A Novel Rule-Based Approach in Mapping Landslide Susceptibility

Sensors ◽  
2019 ◽  
Vol 19 (10) ◽  
pp. 2274 ◽  
Author(s):  
Majid Roodposhti ◽  
Jagannath Aryal ◽  
Biswajeet Pradhan
Keyword(s):  
Landslide Susceptibility ◽  
Area Under The Curve ◽  
Local Scale ◽  
Landslide Susceptibility Mapping ◽  
Susceptibility Map ◽  
Operating Characteristics ◽  
Rule Based ◽  
Conditioning Factors ◽  
Rule Sets ◽  
Rule Based Approach

Despite recent advances in developing landslide susceptibility mapping (LSM) techniques, resultant maps are often not transparent, and susceptibility rules are barely made explicit. This weakens the proper understanding of conditioning criteria involved in shaping landslide events at the local scale. Further, a high level of subjectivity in re-classifying susceptibility scores into various classes often downgrades the quality of those maps. Here, we apply a novel rule-based system as an alternative approach for LSM. Therein, the initially assembled rules relate landslide-conditioning factors within individual rule-sets. This is implemented without the complication of applying logical or relational operators. To achieve this, first, Shannon entropy was employed to assess the priority order of landslide-conditioning factors and the uncertainty of each rule within the corresponding rule-sets. Next, the rule-level uncertainties were mapped and used to asses the reliability of the susceptibility map at the local scale (i.e., at pixel-level). A set of If-Then rules were applied to convert susceptibility values to susceptibility classes, where less level of subjectivity is guaranteed. In a case study of Northwest Tasmania in Australia, the performance of the proposed method was assessed by receiver operating characteristics’ area under the curve (AUC). Our method demonstrated promising performance with AUC of 0.934. This was a result of a transparent rule-based approach, where priorities and state/value of landslide-conditioning factors for each pixel were identified. In addition, the uncertainty of susceptibility rules can be readily accessed, interpreted, and replicated. The achieved results demonstrate that the proposed rule-based method is beneficial to derive insights into LSM processes.

scholarly journals Exploring the potential relationship between the occurrence of landslides and debris flows: A new approach

2020 ◽  
Author(s):  
Zhu Liang ◽  
Changming Wang ◽  
Kaleem Ullah Jan Khan
Keyword(s):  
Random Forest ◽  
Debris Flow ◽  
Debris Flows ◽  
New Approach ◽  
High Class ◽  
Potential Relationship ◽  
Inventory Map ◽  
Auc Value ◽  
Landslides And Debris Flows ◽  
Very High

Abstract. The aim of the present study is to explore the potential relationship between landslides and debris flows by establishing susceptibility zoning maps separately with the use of random forest. Longzi township, Longzi County, located in Southeastern Tibet, where historical landslide and debris flow are commonly occurred, was selected as the study area. The work has been carried out with the following steps: (1) A complete landslide and debris flow inventory map was prepared; (2) Slope units and 11 controlling factors were prepared for the susceptibility modelling of landslide while watershed units and 12 factors for debris flow; (3) Establishing susceptibility zoning maps for landslide and debris flow, respectively, with the use of random forest; (4) The performance of two models are verified using ROC curve, the values of AUC and contingency tables; (5) Putting the high or very-high-class watershed units in the debris flow susceptibility zone map as the base map to observe its coverage by slope units of different classes; (6) The landslide zoning map was put at the bottom floor and analyzed the distribution of high or very-high-class slope units in watershed units; (7) transforming the slope units into points and distributed them on the watershed units. Two models based on random forest have demonstrated great predictive capabilities, of which accuracy was close to 90% and the AUC value was close to 1. The loose sources carried out by the debris flows are not necessarily brought by the landslides although most landslides can be converted into debris flows. The area prone to debris flow does not promote the occurrence of landslides. A susceptibility zoning map composed of two or more natural disasters is comprehensive and significant in this regard.

scholarly journals Identification of the Debris Flow Process Types within Catchments of Beijing Mountainous Area

Water ◽  
2019 ◽  
Vol 11 (4) ◽  
pp. 638 ◽  
Author(s):  
Nan Wang ◽  
Weiming Cheng ◽  
Min Zhao ◽  
Qiangyi Liu ◽  
Jing Wang
Keyword(s):  
Random Forest ◽  
Debris Flow ◽  
Characteristic Curve ◽  
Sediment Concentration ◽  
Mitigation Strategies ◽  
Classification Model ◽  
Gradient Boosting ◽  
Mountainous Area ◽  
Flow Process ◽  
Calculation System

The distinguishable sediment concentration, density, and transport mechanisms characterize the different magnitudes of destruction due to debris flow process (DFP). Identifying the dominating DFP type within a catchment is of paramount importance in determining the efficient delineation and mitigation strategies. However, few studies have focused on the identification of the DFP types (including water-flood, debris-flood, and debris-flow) based on machine learning methods. Therefore, while taking Beijing as the study area, this paper aims to establish an integrated framework for the identification of the DFP types, which consists of an indicator calculation system, imbalance dataset learning (borderline-Synthetic Minority Oversampling Technique (borderline-SMOTE)), and classification model selection (Random Forest (RF), AdaBoost, Gradient Boosting (GBDT)). The classification accuracies of the models were compared and the significance of parameters was then assessed. The results indicate that Random Forest has the highest accuracy (0.752), together with the highest area under the receiver operating characteristic curve (AUROC = 0.73), and the lowest root-mean-square error (RMSE = 0.544). This study confirms that the catchment shape and the relief gradient features benefit the identification of the DFP types. Whereby, the roughness index (RI) and the Relief ratio (Rr) can be used to effectively describe the DFP types. The spatial distribution of the DFP types is analyzed in this paper to provide a reference for diverse practical measures, which are suitable for the particularity of highly destructive catchments.

Sign in / Sign up

Export Citation Format

Share Document