Probability Of Debris Flow Occurrence Research Articles

An updated landslide susceptibility model and a log-Gaussian Cox process extension for Scotland

AbstractAt the time of its development, GeoSure was created using expert knowledge based on a thorough understanding of the engineering geology of the rocks and soils of Great Britain. The ability to use a data-driven methodology to develop a national-scale landslide susceptibility was not possible due to the relatively small size of the landslide inventory at the time. In the intervening 20 years, the National Landslide Database has grown from around 6000 points to over 18,000 records today and continues to be added to. With the availability of this additional inventory, new data-driven solutions could be utilised. Here, we tested a Bernoulli likelihood model to estimate the probability of debris flow occurrence and a log-Gaussian Cox process model to estimate the rate of debris flow occurrence per slope unit. Scotland was selected as the test site for a preliminary experiment, which could potentially be extended to the whole British landscape in the future. Inference techniques for both of these models are applied within a Bayesian framework. The Bayesian framework can work with the two models as additive structures, which allows for the incorporation of spatial and covariate information in a flexible way. The framework also provides uncertainty estimates with model outcomes. We also explored consideration on how to communicate uncertainty estimates together with model predictions in a way that would ensure an integrated framework for master planners to use with ease, even if administrators do not have a specific statistical background. Interestingly, the spatial predictive patterns obtained do not stray away from those of the previous GeoSure methodology, but rigorous numerical modelling now offers objectivity and a much richer predictive description.

Investigation of the functional relationship between antecedent rainfall and the probability of debris flow occurrence in Jiangjia Gully, China

Abstract. A larger antecedent effective precipitation (AEP) indicates a higher probability of a debris flow (Pdf) being triggered by subsequent rainfall. Scientific topics surrounding this qualitative conclusion that can be raised include what kinds of variation rules they follow and whether there is a boundary limit. To answer these questions, Jiangjia Gully in Dongchuan, Yunnan Province, China, is chosen as the study area, and numerical calculation, a rainfall scenario simulation, and the Monte Carlo integration method have been used to calculate the occurrence probability of debris flow under different AEP conditions and derive the functional relationship between Pdf and AEP. The relationship between Pdf and AEP can be quantified by a piecewise function. Pdf is equal to 15.88 %, even when AEP reaches 85 mm, indicating that debris flow by nature has an extremely small probability compared to the rainfall frequency. Data from 1094 rainfall events and 37 historical debris flow events are collected to verify the reasonability of the functional relationship. The results indicate that the piecewise functions are highly correlated with the observation results. Our study confirms the correctness of the qualitative description of the relationship between AEP and Pdf, clarifies that debris flow is a small-probability event compared to rainfall frequency, and quantitatively reveals the evolution law of debris flow occurrence probability with AEP. All the above discoveries can provide a clear reference for the early warning of debris flows.

Effects of rainfall pattern classification methods on the probability estimation of typhoon-induced debris-flow occurrence

ABSTRACT The frequent occurrence of typhoons causes geological disasters, such as debris flow and landslide, by bringing extreme rainfall events. Due to the lack of data collection on extreme rainfall events caused by typhoons, the relationship between rainfall patterns and debris flow has not been deeply studied. Therefore, based on hourly rainfall data during typhoons in Wenzhou from 1980 to 2017, this study used a variety of methods to classify the rainfall events and analyze the characteristics of typhoon-induced rainfall events and their impacts on the probability of debris-flow occurrence. Three classification techniques, including dynamic time warping, K-Means cluster, and self-organizing maps, are applied with two ways to normalize rainfall records, including dimensionless rainfall density curves and dimensionless rainfall cumulation curves, for extracting rainfall patterns from recorded 1 h rainfall data. The rainfall patterns are then used for the estimation of typhoon-induced debris-flow occurrence probability. Results show that different methods present different rainfall patterns. The probability of debris flows varies with different patterns of rainfall events. The research results help deepen the understanding of typhoon rainfall events and debris-flow disaster prevention in the region and contribute to regional flood control and disaster reduction.

Prediction of debris flow initiation points using high resolution DEM and GIS analysis in South Korea

This study presents a GIS analysis model that utilizes high-resolution Digital Elevation Map data to predict the initiation points of debris flow through the overlap of slope stability analysis techniques and rainfall flow analysis. The debris flow that occurred on Mount Woomyun in Seoul, South Korea in 2011 was selected as the case study for comparing and analyzing the prediction model against the actual case. In South Korea, occurrences of debris flow mostly happen during the summer monsoon season. When a substantial amount of rainfall accumulates and heavy rainfall events occur, the pore water pressure within slopes reaches saturation. Consequently, flow is initiated along “temporary streams.” In this study, areas with a combination of low slope stability and abrupt changes in velocity vectors were assumed to have the highest probability of debris flow occurrence due to the overlapping of these factors. The research model was conducted under this assumption. To evaluate the performance of the model, 3D coordinates of the model’s result point, and actual occurrence points were obtained, and a Pearson correlation analysis was conducted to compare them. With these results, R2 values of X = 0.9147, Y = 0.8625, and H = 0.8942 were obtained. These high R2 indicate a strong correlation between the model’s predictions and actual occurrences.

A progressive framework combining unsupervised and optimized supervised learning for debris flow susceptibility assessment

This research aims to improve the reliability of debris flow susceptibility (DFS) assessment, which is crucial for disaster prevention and mitigation in mountainous regions. A progressive framework was proposed and applied to Pinggu District, Beijing, China. First, 16 debris flow predisposing factors (DFPFs) were selected, and the slope structure and soil stability were incorporated to account for material sources. Watershed units instead of grid units were used to extract data. Then, the multi-collinearity among the factors was reduced by using variance inflation factors (VIF) and information gain (IG), and 13 DFPFs were retained. Three unsupervised learning algorithms (i.e., affinity propagation (AP), Gaussian mixture model (GMM) and self-organizing maps (SOM)) were used to optimize the sampling strategy of non-debris flow units. Subsequently, the occurrence probability of debris flows in each unit was predicted by using four supervised learning algorithms (i.e., logistic regression (LR), random forest (RF), adaptive boosting (ADAB) and extreme gradient boosting (XGB)). They were optimized by a state-of-the-art meta-modeling approach. Finally, the DFPF’s importance was ranked. The main contributions of our framework are establishing a high-quality data set and optimizing the prediction algorithms. The results show that the tree-based models perform well, and the boosting-based algorithms outperform the bagging-based algorithms. Supervised learning is more suitable for DFS assessment than unsupervised learning. Debris flows are most likely to occur on a dolomite consequent or diagonal slope with a relief amplitude above 540 m.

Debris flow susceptibility assessment of Leh Valley, Ladakh, based on concepts of connectivity, propagation and evidence-based probability

The Leh Valley which lies within the Trans Himalayan state of Ladakh, India, is known to be affected almost annually by debris flows ranging from minor to catastrophic scale events. The effect has been getting magnified due to increased urbanization and rapid growth in tourism industry. Though these flows are triggered by intense and abnormal rainfall events the conditioning factor has always been the topography and sediment availability. A lucid acknowledgement of the terrain condition and the degree of vulnerability of such events is required. For this a detail investigation of sediment availability, topographic conditions and their relation with known events becomes crucial. This study utilizes index of connectivity (IC) model to understand the sediment source-sink relationship and farther applied Flow-R model to simulate the probable scenario of events through predefined algorithms. We then use the Weights of evidence (WOE) method to compute the statistical probability of debris flow occurrence. This paper demonstrates the application of these three independent techniques and their implementation in a highly rugged terrain of Ladakh which is a region of frequent debris flows onslaught. The IC and Flow-R models are found to be counter supportive and effective in delineating areas which could be affected by flows that will solely originate in upstream areas where high angle channels directly connected to sediment sources are present. WOE-based model determines the probability of the rare and extensive flows that results from downward integration of other drainage networks in an open fan area.

Evaluation of Rainfall-Triggered Debris Flows under the Impact of Extreme Events: A Chenyulan Watershed Case Study, Taiwan

This study examined the conditions that lead to debris flows, and their association with the rainfall return period (T) and the probability of debris flow occurrence (P) in the Chenyulan watershed, central Taiwan. Several extreme events have occurred in the Chenyulan watershed in the past, including the Chi-Chi earthquake and extreme rainfall events. The T for three rainfall indexes (i.e., the maximum hourly rainfall depth (Im), the maximum 24-h rainfall amount (Rd), and RI (RI = Im× Rd)) were analyzed, and the T associated with the triggering of debris flows is presented. The P–T relationship can be determined using three indexes, Im, Rd, and RI; how it is affected and unaffected by extreme events was developed. Models for evaluating P using the three rainfall indexes were proposed and used to evaluate P between 2009 and 2020 (i.e., after the extreme rainfall event of Typhoon Morakot in 2009). The results of this study showed that the P‒T relationship, using the RI or Rd index, was reasonable for predicting the probability of debris flow occurrence.

Uncertainty analysis of a rainfall threshold estimate for stony debris flow based on the backward dynamical approach

Abstract. A rainfall threshold is a function of some rainfall quantities that provides the conditions beyond which the probability of debris-flow occurrence is considered significant. Many uncertainties may affect the thresholds calibration and, consequently, its robustness. This study aims to assess the uncertainty in the estimate of a rainfall threshold for stony debris flow based on the backward dynamical approach, an innovative method to compute the rainfall duration and averaged intensity strictly related to a measured debris flow. The uncertainty analysis is computed by performing two Monte Carlo cascade simulations: (i) to assess the variability in the event characteristics estimate due to the uncertainty in the backward dynamical approach parameters and data and (ii) to quantify the impact of this variability on the threshold calibration. The application of this procedure to a case study highlights that the variability in the event characteristics can be both low and high. Instead, the threshold coefficients have a low dispersion showing good robustness of the threshold estimate. Moreover, the results suggest that some event features are correlated with the variability of the rainfall event duration and intensity. The proposed method is suitable to analyse the uncertainty of other threshold calibration approaches.

A method for ex-post benefit-cost assessment for engineering remediation of debris flow impacts

Ex-ante and ex-post assessments are both essential in risk management, although ex-post assessments of engineering measures have not been widely reported. In this study, an ex-post benefit-cost assessment method was developed to enable the direct impact of remediation engineering measures for debris flows to be quantified. We propose three categories of parameters for calculating benefit and cost: (1) predicted benefits (comprising predicted benefits of saving lives, at the household level, and of infrastructure protection); (2) actual loss (of losing lives, at the household level, and of infrastructure); (3) actual cost of the engineering. Actual benefit was defined as predicted benefit minus loss after completion of the engineering measure. The engineering measures were regarded as effective when benefits outweighed costs. Government data, survey results based on past disasters, and interview results with specialists were used. This methodology was applied and tested in ten villages located in the Gaoping River Basin in southern Taiwan. The result showed that the range of the benefit and cost ratio was between 1.26 and 7.99. We recommend this ex-post assessment method for rapid screening of districts with low benefit-cost ratios. • An ex post benefit–cost assessment method for remediation engineering measures of debris flows. • Predicted benefits, actual loss, and actual cost used to evaluate benefits and costs. • Probability of debris flow occurrence after an alert warning included in the predicted benefits. • The method considers both losses of remediation engineering measures and damages at household level.

Topographical and geological variation of effective rainfall for debris-flow occurrence from a large-scale perspective

Understanding the spatial heterogeneity of the debris-flow rainfall threshold is meaningful for regional warning systems. However, regional topographical and geological impacts have not been well quantified. Using the Hengduan Mountains with distinct climatic, topographical, and geological settings as a typical study area, this study focused on the geological and topographical variations of effective rainfall for debris-flow occurrence from a large-scale perspective. Results showed that the probability of debris-flow occurrence was smallest in the cold-arid zone and largest in the humid zone. The mean dimensionless effective rainfall (the ratio of effective rainfall to mean annual precipitation) was about 5.2% for an initial warning and 7.7% for serious warnings. The contributions to the differences are regional climate, geomorphology, and geology, in sequence of importance. The zone with more complex terrain, thicker soil, and higher sand content would be more beneficial to debris-flow formation. Multiple regression analysis suggested that the geologically-related aspects may have a slightly greater effect than topographical factors on effective rainfall. In general, this work could be in favor of other similar regional, national, and even global work for debris-flow early warning.

A Logistic Model Predicting Occurrence Probability of Debris Flow

This paper presents a logistic model for predicting the occurrence probability of debris flows based on rainfall intensity and duration. The data from a total of 354 rainfall events were used to calibrate the model, among which 249 were triggering a debris flow while 105 were not. The model will be useful to the decision making of debris flow early warning in the future. That is, given the estimated occurrence probability = 70% subject to a combination of rainfall intensity and duration, there is a 30% probability that the early warning will be a false alarm. By contrast, if decision makers decide not to issue an early warning, then there is a 70% chance leading to a missed alarm. Subsequently, integrating the consequences of missed alarm and false alarm into the equation, the respective risks can be computed, based on which decision makers can make a more robust decision whether an early warning is needed or not by choosing the scenario with a lower risk.

Investigation of Post-Fire Debris Flows in Montecito

Debris flows in a burned area, post-fire debris flows, are considered as one of the most dangerous geo-hazards due to their high velocity, long run-out distance, and huge destruction to infrastructures. The rainfall threshold to trigger such hazards is often reduced compared with normal debris flow because ashes generated by mountain fires reduce the permeability of the top soil layer, thus increasing surface runoff. At the same time, burnt material and residual debris have very poor geo-mechanical characteristics, e.g., their internal friction angle and cohesion are typically low, and thus an intense rainfall can easily trigger some debris flows. Studying post-fire debris flow enables us to get a deeper understanding of disaster management. In this paper, the debris flow that occurred in Montecito, California, USA, and was affected by the Thomas Fire was used as a case study. Five major watersheds were extracted based on the digital elevation model (DEM). Remote sensing images were used to analyze the wildfire process, the extent of the burned areas, and the burn severity. The hypsometric integral (HI) and short-duration rainfall records of the watersheds around Montecito when the post-fire debris flows occurred were analyzed. Steep terrain, loose and abundant deposits, and sufficient water supply are the important conditions affecting the formation of debris flows. Taking watersheds as the research objects, HI was used to describe the geomorphic and topographic features, open-access rainfall data was used to represent the water supply, and burn severity represented the abundance of material sources. An occurrence probability model of post-fire debris flow based on HI, short-duration heavy rainfall, and burn severity was developed by using a logistic regression model in post-fire areas. By using this model, the occurrence probability of the post-fire debris flow in different watersheds around Montecito was analyzed based on the precipitation with time. Especially, the change characteristics of occurrence probability of debris flows over time based on the model bring a new perspective to observe the obvious change of the danger of post-fire debris flows and it is very useful for early warning of post-fire debris flows.

Effects of landslide and other physiographic factors on the occurrence probability of debris flows in central Taiwan

This study focuses on the effects of physiographic factors, such as landslide area, on the occurrence of debris flows in central Taiwan. Four physiographic factors were selected for their significance to the occurrence of debris flows, including landslide ratio (the ratio of landslide area over watershed area), average steepness of the streambed, effective watershed area, and form factor. Two quantifying methods of factors were performed and compared: one using genuine values of factors and another one using values converted by degree of membership from fuzzy theory. Then the logistic regression method was applied for building a model to assess the occurrence probability of debris flows from five variables: four physiographic factors and one hydrologic factor. The model is consistent with the mechanism of debris flow occurrence, with all physiographic and hydrologic factors positively correlated with the occurrence probability. In addition, the accuracy of the model was validated with randomly selected historical events and demonstrated fairly satisfactory validity, ranging from 70 to 80 %. It was found that adopting the degree of membership made the model more stable and more reliable. In addition, the model also shows that reducing the landslide area can significantly reduce the occurrence probability of debris flows. The results show that the model built in this study has the potential to be well applied and fully integrated into current or future warning systems.

Risk Assessment of the Debris Flow Gully in Tibet Southeast

Risk assessment of debris flow is the core content and decision-making basis for debris flow disaster forecasting and the disaster prevention work. It is a comprehensive analysis for the geological, climate, rainfall, historical disaster activities and human engineering activities in some certain areas which may be in danger conditions, so as to determine the occurrence probability of debris flow. In this study, risk assessment model of the debris flow gully in Tibet southeast area is established. According to the correlation ratio of risk assessment indexes and the debris flow occurrence, the assessment indexes can be divided into three levels and defined with some certain values according to the correlation of debris flow occurrence. Finally, the risk assessment of debris flow gully in Tibet southeast area is conducted, and model outputs achieve a good result.

Debris flow susceptibility assessment using a probabilistic approach: A case study in the Longchi area, Sichuan province, China

The Longchi area with the city of Dujiangyan, in the Sichuan province of China, is composed of Permian stone and diorites and Triassic sandstones and mudstones intercalated with slates. An abundance of loose co-seismic materials were present on the slopes after the May 12, 2008 Wenchuan earthquake, which in later years served as source material for rainfall-induced debris flows or shallow landslides. A total of 48 debris flows, all triggered by heavy rainfall on 13 th August 2010, are described in this paper. Field investigation, supported by remote sensing image interpretation, was conducted to interpret the co-seismic landslides in the debris flow gullies. Specific characteristics of the study area such as slope, aspect, elevation, channel gradient, lithology, and gully density were selected for the evaluation of debris flow susceptibility. A score was given to all the debris flow gullies based on the probability of debris flow occurrence for the selected factors. In order to get the contribution of the different factors, principal component analyses were applied. A comprehensive score was obtained for the 48 debris flow gullies which enabled us to make a susceptibility map for debris flows with three classes. Twenty-two gullies have a high susceptibility, twenty gullies show a moderate susceptibility and six gullies have a low susceptibility for debris flows.

The DEFENSE (debris Flows triggEred by storms – nowcasting system): An early warning system for torrential processes by radar storm tracking using a Geographic Information System (GIS)

Debris flows, responsible for economic losses and occasionally casualties in the alpine region, are mainly triggered by heavy rains characterized by hourly peaks of varying intensity, depending on the features of the basin under consideration. By integrating a recent classification of alpine basins with the radar storm tracking method, an innovative early warning system called DEFENSE (DEbris Flows triggEred by storms – Nowcasting SystEm) was developed using a Geographical Information System (GIS). Alpine catchments were classified into three main classes based on the weathering capacity of the bedrock into clay or clay-like minerals, the amount of which, in unconsolidated material, directly influences the debris flow rheology, and thus the sedimentary processes, the alluvial fan architecture, as well as the triggering frequency and seasonal occurrence probability of debris flows. Storms were identified and tracked by processing weather radar observations; subsequently, rainfall intensities and storm severity were estimated over each classified basin. Due to rainfall threshold values determined for each basin class, based on statistical analysis of historical records, an automatic corresponding warning could be issued to municipalities.

A prediction model for debris flows triggered by a runoff-induced mechanism

Many debris flows were triggered within and also outside the Dayi area of the Guizhou Province, China, during a rainstorm in 2011. High-intensity short-duration rainfall was the main triggering factor for these gully-type debris flows which are probably triggered by a runoff-induced mechanism. A revised prediction model was introduced for this kind of gully-type debris flows with factors related to topography, geology, and hydrology (rainfall) and applied to the Wangmo River catchment. Regarding the geological factor, the “soft lithology” and “loose sediments” in the channel were added to the list of the average firmness coefficient for the lithology. Also, the chemical weathering was taken into account for the revised geological factor. Concerning the hydrological factor, a coefficient of variation of rainfall was introduced for the normalization of the rainfall factor. The prediction model for debris flows proposed in this paper delivered three classes of the probability of debris flow occurrence. The model was successfully validated in debris flow gullies with the same initiation mechanism in other areas of southwest China. The generic character of the model is explained by the fact that its factors are partly based on the initiation mechanisms and not only on the statistical analyses of a unique variety of local factors. The research provides a new way to predict the occurrence of debris flows initiated by a runoff-induced mechanism.

Debris-flow hazard assessment and validation following the Medano Fire, Great Sand Dunes National Park and Preserve, Colorado

A debris-flow hazard assessment was conducted for the Medano Creek drainage basin, at Great Sand Dunes National Park and Preserve, following the 2,400 ha Medano Fire in 2010. Debris-flow probability and volume predictions were made using empirical regression models. Model parameters include burn severity, rainfall intensity, topographic characteristics, and soil properties. Model results provided park resource managers with information on potential sub-basin-specific hazards to park visitors, roads, and campsites. Monitoring of the first significant rainfall events following the fire and the resulting debris-flow responses provided means for model validation. Of the three probability models utilized, two predicted high probability of debris-flow occurrence for all sub-basins that produced debris flows, with numerous false positives, while the third failed to predict high probability in any of the sub-basins. The volume model predicted volumes within approximately one order of magnitude higher than those measured.

An empirical approach for evaluation of the potential of debris flow occurrence in mountainous areas

Debris flows can occur relatively suddenly and quickly in mountainous areas, resulting in major structural damage and loss of life. The establishment of a model to evaluate the occurrence probability of debris flows in mountainous areas is therefore of great value. The influence factors of debris flows are very complex; they can basically be divided into background factors and triggering factors. Background factors include the mechanical characteristics of geo-materials, topography and landscape, and soil vegetation; and triggering factors include hydrological and rainfall conditions, and human activities. By assessing the dynamic characteristics of debris flows in mountainous areas, some important influence factors are selected here for analysis of their impacts on the occurrence probability of debris flow. A mathematical model for evaluation of the occurrence probability of debris flows is presented and combined with probability analysis. Matlab software is used for the numerical implementation of the forecasting model, and the influences of rainfall, lithology and terrain conditions on the occurrence probability of debris flows are analyzed. Finally, the presented model is applied to forecast the occurrence probability of debris flows in the mountainous area around Qingping Town; the simulation results show that many loose landslide deposits and heavy rainfall are the key factors likely to trigger debris flows in this region.

Relationships between debris flows and earth surface factors in Southwest China

Southwest China, including the Provinces of Yunnan, Guizhou, Sichuan and Chongqing, is a region with serious debris flow hazards, where 7,561 debris flow sites have been identified. Based on the data from these sites, the distribution regularity of debris flows was analyzed. Earth surface factors that may influence the formation of debris flows were analyzed from the viewpoints of energy and material conditions. Four major earth surface factors were selected: relative relief, stratigraphy, fault density and land-use conditions. With the support of GIS, the research region was divided into 125,177 grid cells and for each cell data for the four factors were collected. Based on this information, the distribution of quantity and the occurrence probability of debris flows and the role of each factor were statistically analyzed. The results should be helpful for the assessment of debris flow hazards and debris flow forecasting in the research region.