Debris Flow Susceptibility Mapping Research Articles

Comparison of Different Negative-Sample Acquisition Strategies Considering Sample Representation Forms for Debris Flow Susceptibility Mapping

The lack of reliable negative samples is an important factor limiting the quality of machine learning-based debris flow susceptibility mapping (DFSM). The purpose of this paper is to propose multiple negative-sample acquisition strategies for DFSM considering different sample representation forms. The sample representation forms mainly include a single grid, multi-grid, and watershed unit, and the negative-sample acquisition strategies are based on support vector machine (SVM), spy technique, and isolation forest (IF) methods, respectively. These three strategies can assign a value to all the samples based on different assumptions, and reliable, negative samples can be generated from samples with values below a predefined threshold. Combining different sample representation forms with negative sample acquisition strategies, nine datasets were then involved in random forest (RF) modeling. The receiver operating characteristic (ROC) curves and related statistical results were used to evaluate the models. The results show that the strategy based on the spy technique is suitable for multiple datasets, while the IF-based strategy is well-adapted to the watershed unit datasets. This study can provide more options for improving the quality of datasets in DFSM, which can further improve the performance of machine learning models.

A debris flow susceptibility mapping study considering sample heterogeneity

A debris flow susceptibility mapping study considering sample heterogeneity

Dependence of debris flow susceptibility maps on sampling strategy with data-driven grid-based model

Different sampling strategies produce varying sample data, serve as the primary input data and directly affect the accuracy of predictions in data-driven grid-based susceptibility models. This study analyzes the accuracy and variation of debris flow susceptibility maps (DFSMs) generated by various sampling strategies. The study area is the Yingxiu region in China, where six sampling strategies were applied, including three sampling locations (deposition area, runout area, and source area) and two sampling types (centroid and polygon) for the debris flow inventory. The effectiveness of 10 conditioning factors used to build the model was assessed by using Pearson correlation coefficient, variance inflation factor, and information gain ratio (IGR) techniques. We then used Weight of Evidence (WofE), Logistic Regression (LR), and Deep Neural Network (DNN) models to produce DFSMs and quantify their performance using the receiver operating characteristic curve (ROC), Accuracy (ACC), Precision, F1 score, and Recall. The results show that the WofE (AUC: 0.754–0.960), LR (AUC: 0.761–0.965), and DNN (AUC: 0.786–0.976) models all perform well, but the DFSMs and dominant factors depend strongly on sampling strategies, especially on sampling location. If the sample areas are excessively large and span across different factor class labels, or if there is a concentration of either large or small sample areas within a specific region, the results of centroid and polygon sampling strategies may differ or even be contradictory.We recommend: (1) determining sampling locations based on the research objectives to provide more accurate evaluation results; (2) selecting the sampling type by first considering the sample size. If the aforementioned conditions are not present, the quicker and more convenient centroid sampling strategy can be chosen; and (3) determining an appropriate sampling strategy and ensuring the accuracy of initial samples are paramount before producing DFSMs.

Debris flow susceptibility mapping in alpine canyon region: a case study of Nujiang Prefecture

Debris flow susceptibility mapping in alpine canyon region: a case study of Nujiang Prefecture

Glacial debris flow susceptibility mapping based on combined models in the Parlung Tsangpo Basin, China

Glacial debris flow susceptibility mapping based on combined models in the Parlung Tsangpo Basin, China

Application of different watershed units to debris flow susceptibility mapping: A case study of Northeast China

The main purpose of this study was to compare two types of watershed units divided by the hydrological analysis method (HWUs) and mean curvature method (CWUs) for debris flow susceptibility mapping (DFSM) in Northeast China. Firstly, a debris flow inventory map consisting of 129 debris flows and 129 non-debris flows was randomly divided into a ratio of 70% and 30% for training and testing. Secondly, 13 influencing factors were selected and the correlations between these factors and the debris flows were determined by frequency ration analysis. Then, two types of watershed units (HWUs and CWUs) were divided and logistic regression (LR), multilayer perceptron (MLP), classification and regression tree (CART) and Bayesian network (BN) were selected as the evaluation models. Finally, the predictive capabilities of the models were verified using the predictive accuracy (ACC), the Kappa coefficient and the area under the receiver operating characteristic curve (AUC). The mean AUC, ACC and Kappa of four models (LR, MLP, CART and BN) in the training stage were 0.977, 0.931, and 0.861, respectively, for the HWUs, while 0.961, 0.905, and 0.810, respectively, for the CWUs; in the testing stage, were 0.904, 0.818, and 0.635, respectively, for the HWUs, while 0.883, 0.800, and 0.601, respectively, for the CWUs, which showed that HWU model has a higher debris flow prediction performance compared with the CWU model. The CWU-based model can reflect the spatial distribution probability of debris flows in the study area overall and can be used as an alternative model.

Impact of LULC on debris flow using linear aggression model from Gilgit to Khunjerab with emphasis on urban sprawl.

In this research, the impact of land use and land cover (LULC) on debris flow was evaluated in the Gilgit to Khunjerab region. Two events have been done: (i) LULC stimulations for 2026 and 2030 using the MOLUSCE plugin and (ii) debris flow susceptibility mapping using linear aggression model. The evaluation of LULC on debris flow susceptibility is based on two scenarios: (i) existing (2010, 2014, 2018, 2022) LULC scenarios and (ii) stimulated (2026, 2030) LULC scenarios. The linear aggression model has 16 contributing factors to developing the debris flow susceptibility mapping. The main contributing components in debris flow susceptibility mapping are slope and LCCS. According to the linear aggressiveness model, debris flow susceptibility grows as the LULC changes, and the high susceptibility zones' share increases. For the current years 2010, 2014, 2018, and 2022, as well as the stimulated years 2026 and 2030, the model had high success rates (> 90.0%) and prediction rates (> 85.0%). The findings backed up prior research and suggested that the impact of LULC will grow in the future.

Debris flow susceptibility mapping in Colorado Front Range, USA: A comparison of physics-based and data-driven approaches

This study used an inventory of debris flow triggered by a storm event in Colorado Front Range as an example to compare the capability of data-driven and physics-based approaches for regional-scale debris flow susceptibility mapping (LSM). Nine debris flow contributing factors were collected for the present study based on the availability of geophysical data in the study area. These contributing factors represent hillslope geometries, surface hydrology, and soil conditions. For the physics-based approach, the infinite slope model was used to directly determine the debris flow susceptibility for the study area by calculating the factor of safety (FS) based on parameters derived from geophysical data. For the data-driven approach, an artificial neural network (ANN) was developed to predict debris flow susceptibility for the study area by learning relationships from the contributing factors using the debris flow inventory. The results showed that both physics-based and data-driven models predicted debris flow susceptibility in the study area with relatively high accuracy; the data-driven approach outperformed the physics-based approach as it could extract complex features which the physics-based approach did not consider.

Debris flow susceptibility mapping using the Rock Engineering System (RES) method: a case study

The main purpose of the present study is to develop a debris flow susceptibility map of a mountain area (Susa Valley, Western Italian Alps) by using an upgraded version of the Bonetto et al. (Journal of Mountain Science 18, 2021) approach based on the Rock Engineering System (RES) method. In particular, the area under investigation was discretized in a 5 × 5-m grid on which GIS-based analyses were performed. Starting from available databases, several geological, geo-structural, morphological and hydrographical predisposing parameters were identified and codified into two interaction matrices (one for outcropping lithologies and one for Quaternary deposits), to evaluate their mutual interactions and their weight in the susceptibility estimation. The result for each grid point is the debris flow propensity index (DfPI), an index that estimates the susceptibility of the cell to be a potential debris flow source. The debris flow susceptibility map obtained was compared with those obtained from two expedited and universally recognized susceptibility methods, i.e. the Regional Qualitative Heuristic Susceptibility Mapping (RQHSM) and the Likelihood Ratio (LR). Each map was validated by using the Prediction Rate Curve method. The limitations and strong points of the approaches analysed are discussed, with a focus on the innovativeness and uniqueness of the RES. In fact, in the study site, the RES method was the most efficient for the detection of potential source areas. These results prove its robustness, cost-effectiveness and speed of application in the identification and mapping of sectors capable of triggering debris flow.

Debris flow susceptibility mapping in mountainous area based on multi-source data fusion and CNN model – taking Nujiang Prefecture, China as an example

ABSTRACT Efforts to evaluate the susceptibility of debris flows in large areas, especially in mountainous regions, are often hampered by the alpine and canyon terrain. This paper proposes a convolution neural network (CNN) model named dense residual shuffle net (DRSNet). It is successfully applied to Nujiang Prefecture in Yunnan Province of China, a typical alpine area with frequent debris flows. DRSNet uses digital elevation model, remote sensing, lithology, soil type and precipitation data as input. First, dense connection and residual structure were used to extract the shallow features of various data. Next, channel shuffle, fuse block and fully connection were applied to strengthen the correlation between different shallow features and give inner danger scores. Finally, precipitation as the activation factor was introduced giving the valleys susceptibility. To verify the feasibility of DRSNet, comparative tests were conducted on 7 CNN models and 3 other machine learning (ML) methods. Experimental results show that DRSNet can achieve 78.6% accuracy in debris flow valley classification, which is at least 7.4% higher than common CNN models and 15.2% higher than other ML methods. This article provides new ideas for debris flow susceptibility evaluation.

A Research on Cross-Regional Debris Flow Susceptibility Mapping Based on Transfer Learning

Debris flow susceptibility mapping (DFSM), which has proven to be one of the most effective tools for risk management, faces a variety of problems. To realize the rational use of debris flow sample resources and improve the modeling efficiency, a unified model based on transfer learning was established for cross-regional DFSM. First, samples with 10 features collected from two debris flow-prone areas were separately used to perform factor prediction ability analysis (FPAA) based on the information gain ratio (IGR) method and then develop traditional machine learning models based on random forests (RF). Secondly, two feature matrices representing different areas were projected into a common latent feature space to obtain two new feature matrices. Then, the samples with new features were used together for FPAA and developing a unified machine learning model. Finally, the performance of the models was obtained and compared based on the area under curves (AUC) and some statistical results. All the conditioning factors played different roles in debris flow prediction in the two study areas, based on which two traditional models and a unified model were established. The unified model based on feature transferring realized efficient cross-regional modeling, solved the unconvincing problem of limited sample modeling, and enabled more accurate identification of some debris flow samples.

Debris-Flow Susceptibility Assessment in China: A Comparison between Traditional Statistical and Machine Learning Methods

Debris flows, triggered by dual interferences extrinsically and intrinsically, have been widespread in China. The debris-flow susceptibility (DFS) assessment is acknowledged as the benchmark for the mitigation and prevention of debris flow risks, but DFS assessments at the national level are lacking. The role of human activities in the DFS assessment has always been overlooked. On the basis of a detailed inventory of debris-flow sites and a large set of environmental and human-related characteristics, this research presents the comparative performance of the well-known information value (IV), logistic regression (LR) and random forest (RF) models for DFS assessments in China. Twelve causative factors, namely, elevation, slope, aspect, rainfall, the normalized difference vegetation index (NDVI), land use, landform, geology, distance to faults, density of villages, distance to rivers and distance to roads, were considered. Debris-flow susceptibility maps were then generated after the nonlinear relationship between the debris-flow occurrence and the causative factors was captured. Finally, the predictive performance of the three maps was evaluated through receiver operating characteristic (ROC) curves, and the validation results showed that areas under the ROC curves were 81.98%, 79.96% and 97.38% for the IV, LR and RF models, respectively, indicating that the RF model outperformed the other two traditional statistical methods. The importance ranking of the RF model also revealed that distance to roads, slope and rainfall dominated the spatial distribution of debris flows. This is the first experiment to compare between the traditional statistical and machine learning methods in DFS studies for the whole of China. Our results could provide some empirical support for China’s policymakers and local practitioners in their efforts to enable residents to be less vulnerable to disasters.

Establishing a GIS-based evaluation method considering spatial heterogeneity for debris flow susceptibility mapping at the regional scale

Establishing a GIS-based evaluation method considering spatial heterogeneity for debris flow susceptibility mapping at the regional scale

Susceptibility Mapping of Typical Geological Hazards in Helong City Affected by Volcanic Activity of Changbai Mountain, Northeastern China

The purpose of this paper was to produce the geological hazard-susceptibility map for the Changbai Mountain area affected by volcanic activity. First, 159 landslides and 72 debris flows were mapped in the Helong city are based on the geological disaster investigation and regionalization (1:50,000) project of Helong City. Then, twelve landslide conditioning factors and eleven debris flow conditioning factors were selected as the modeling variables. Among them, the transcendental probability of Changbai Mountain volcanic earthquake greater than VI degrees was used to indicate the relationship between the geological hazard-susceptibility and Changbai Mountain volcanic earthquake occurrence. Furthermore, two machine learning models (SVM and ANN) were introduced to geological hazard-susceptibility modeling. Receiver operating characteristic curve, statistical analysis method, and five-fold cross-validation were used to compare the two models. Based on the modeling results, the SVM model is the better model for both the landslide and debris flow susceptibility mapping. The results show that the areas with low, moderate, high, and very high landslide susceptibility are 31.58%, 33.15%, 17.07%, and 18.19%, respectively; and the areas with low, moderate, high, and very high debris flow susceptibility are 25.63%, 38.19%, 23.47%, and 12.71%, respectively. The high and very high landslide and debris flow susceptibility classes make up 85.54% and 80.55% of the known landslides and debris flow, respectively. Moreover, the very high and high landslide and debris flow susceptibility are mainly distributed in the lower elevation area, and mainly distributed around the cities and towns in Helong City. Consequently, this paper will be a useful guide for the deployment of disaster prevention and mitigation in Helong city, and can also provide some reference for evaluation of landslide susceptibility in other volcanically active areas.

Global debris flow susceptibility based on a comparative analysis of a single global model versus a continent-by-continent approach

Debris flows, and landslides in general, are worldwide catastrophic phenomena. As world population and urbanization grow in magnitude and geographic coverage, the need exists to extend focus, research, and modeling to a continental and global scale. Although debris flow behavior and parameters are local phenomena, sound generalizations can be applied to debris flow susceptibility analyses at larger geographic extents based on these criteria. The focus of this research is to develop a global debris flow susceptibility map by modeling at both a continental scale for all continents and by a single global model and determine whether a global model adequately represents each continent. Probability Density, Conditional Probability, Certainty Factor, Frequency Ratio, and Maximum Entropy statistical models were developed and evaluated for best model performance using fourteen environmental factors generally accepted as the most appropriate debris flow predisposing factors. Global models and models for each continent were then developed and evaluated against verification data. The comparative analysis demonstrates that a single global model performs comparably or better than individual continental models for a majority of the continents, resulting in a debris flow susceptibility map of the world useful for in-depth research, international planning, and future debris flow susceptibility modeling and determining societal impacts.

Morphometry and Debris-Flow Susceptibility Map in Mountain Drainage Basins of the Vallo di Diano, Southern Italy

In watershed mountain basins, affected in the last decades by strong rainfall events, the role of debris-flow and debris flood processes was investigated. Morphometric parameters have proven to be useful first-approximation indicators in discriminating those processes, especially in large areas of investigation. Computation of morphometric parameters in 19 watershed mountain basins of the western side valley of the Vallo di Diano intermontane basin (southern Italy) was carried out. This procedure was integrated by a semi-automatic elaboration of the potential susceptibility to debris flows, using Flow-R modelling. This software, providing an empirical model of the preliminary susceptibility assessment at a regional scale, was applied in many countries of the world. The implementation of Flow-R modelling requires a GIS application and some thematic base maps extracted using DEMs analysis. A 5-meter-resolution DEM has been used in order to produce the susceptibility maps of the whole study area, and the results are compared and discussed with the real debris flow/flood events that occurred in 1993, 2005, 2010, and 2017 in the studied area. The results have provided a good reliability of Flow-R modelling within small catchment mountain basins.

Comparison of different sampling strategies for debris flow susceptibility mapping: A case study using the centroids of the scarp area, flowing area and accumulation area of debris flow watersheds

Comparison of different sampling strategies for debris flow susceptibility mapping: A case study using the centroids of the scarp area, flowing area and accumulation area of debris flow watersheds

Development of a region-partitioning method for debris flow susceptibility mapping

Debris flow susceptibility mapping (DFSM) has been reported in many studies, however, the irrational use of the same conditioning factor system for DFSM in regional-scale has not been thoroughly resolved. In this paper, a region-partitioning method that is based on the topographic characteristics of watershed units was developed with the objective of establishing multiple conditioning factor systems for regional-scale DFSM. First, watershed units were selected as the mapping units and created throughout the entire research area. Four topographical factors, namely, elevation, slope, aspect and relative height difference, were selected as the basis for clustering watershed units. The k-means clustering analysis was used to cluster the watershed units according to their topographic characteristics to partition the study area into several parts. Then, the information gain ratio method was used to filter out superfluous factors to establish conditioning factor systems in each region for the subsequent debris flow susceptibility modeling. Last, a debris flow susceptibility map of the whole study area was acquired by merging the maps from all parts. DFSM of Yongji County in Jilin Province, China was selected as a case study, and the analytical hierarchy process method was used to conduct a comparative analysis to evaluate the performance of the region-partitioning method. The area under curve (AUC) values showed that the partitioning of the study area into two parts improved the prediction rate from 0.812 to 0.916. The results demonstrate that the region-partitioning method on the basis of topographic characteristics of watershed units can realize more reasonable regional-scale DFSM. Hence, the developed region-partitioning method can be used as a guide for regional-scale DFSM to mitigate the imminent debris flow risk.

Debris flows modeling using geo-environmental factors: developing hybridized deep-learning algorithms

Although the prediction of debris flow-prone areas represents a key step towards reducing damages, modeling debris flow susceptibility is complicated. In addition, the role of debris flow causal drivers in forested mountain landscapes is still poorly understood. To gain a holistic view of the causes of debris flows in the Umyeonsan, Seoul, South Korea region, we coupled the convolutional neural network (CNN) with two evolutionary optimization algorithms – grey wolf optimization (GWO) and cuckoo optimization algorithm (COA). Applying geoinformatics to debris flow factors, debris-flow susceptibility maps were generated and their validities were assessed with receiver operating characteristic (ROC) curves. The results reveal that three causative factors seem to contribute most to debris flows in the study area. The evolutionary optimization algorithms achieved higher goodness-of-fit and predictive power than the standalone CNN model. The goodness-of-fit and predictive skill measures of the CNN susceptibility map were 0.76 and 0.73. The values of CNN hybridized with GWO were 0.81 and 0.81 and hybridized with COA were 0.83 and 0.82. Slope degree, tree age, stream power index, geographical class, and soil drainage were the factors most affecting debris flow likelihood. The CNN-COA is the superior model and it predicted that 40.6% of the study area (i.e., 1844.96 km2) is highly and very highly susceptible to debris flows. The methodology can be applied for analysis of other region to improve risk management and guide development and land use planning.

Mapping Susceptibility to Debris Flows Triggered by Tropical Storms: A Case Study of the San Vicente Volcano Area (El Salvador, CA)

In this study, an inventory of storm-triggered debris flows performed in the area of the San Vicente volcano (El Salvador, CA) was used to calibrate predictive models and prepare a landslide susceptibility map. The storm event struck the area in November 2009 as the result of the simultaneous action of low-pressure system 96E and Hurricane Ida. Multivariate Adaptive Regression Splines (MARS) was employed to model the relationships between a set of environmental variables and the locations of the debris flows. Validation of the models was performed by splitting 100 random samples of event and non-event 10 m pixels into training and test subsets. The validation results revealed an excellent (area under the receiver operating characteristic (ROC) curve (AUC) = 0.80) and stable (AUC std. dev. = 0.01) ability of MARS to predict the locations of the debris flows which occurred in the study area. However, when using the Youden’s index as probability threshold to discriminate between pixels predicted as positives and negatives, MARS exhibits a moderate ability to identify stable cells (specificity = 0.66). The final debris flow susceptibility map, which was prepared by averaging for each pixel the score of the 100 MARS repetitions, shows where future debris flows are more likely to occur, and thus may help in mitigating the risk associated with these landslides.