scholarly journals Forecasting of Debris Flow Using Machine Learning-Based Adjusted Rainfall Information and RAMMS Model

Water ◽  
2021 ◽  
Vol 13 (17) ◽  
pp. 2360
Author(s):  
Cheong-Hyeon Oh ◽  
Kyung-Su Choo ◽  
Chul-Min Go ◽  
Jung-Ryel Choi ◽  
Byung-Sik Kim
Keyword(s):  
Machine Learning ◽  
Debris Flow ◽  
Debris Flows ◽  
Precipitation Forecast ◽  
Main Body ◽  
Observation Data ◽  
Flow Models ◽  
The Past ◽  
Quantitative Precipitation Forecast ◽  
Heavy Rainfalls

In recent years, climate change and extreme weather conditions have caused natural disasters of various sizes and forms across the world. The increase in the resulting flood damage and secondary damage has also inflicted massive social and economic harm. Korea is no exception, where debris flows created by typhoons and localized heavy rainfalls have caused human injuries and property damage in the Wumyeonsan Mountain in Seoul, Majeoksan Mountain in Chuncheon, Sinnam in Samcheok, Gokseong in Jeollanam-do, and Anseong in Gyeonggi-do. Disaster damage needs to be minimized by preparing for typhoons and heavy rainfalls that cause debris flow. To that end, we need accurate prediction of rainfall and flooding through simulations based on debris flow models. Most of the previous literature analyzed debris flows using rainfall events in the past before debris flow occurrence, rather than analyzing and predicting based on rainfall predictions. The main body of this study assesses the applicability of hydrological quantitative precipitation forecast (HQPF) generated through a machine learning method named the Random Forest (RF) method to debris flow analysis models. To that end, this study uses scatter plots to compare and analyze the precipitation observation data collected from the areas hit by debris flows in the past, and the quantitative precipitation forecast (QPF) and HQPF data from the Korea Meteorological Administration (KMA). Based on the verified HQPF data, runoff was calculated using the spatial runoff assessment tool (S-RAT) model, and the soil amount was calculated to simulate the debris flow damage with a two-dimensional rapid mass movements (RAMMS) model. The debris flow simulation based on the said data indicated varying degrees of flow depth, impact force, speed, and damage area depending on the precipitation. The correction of the HQPF was verified by measuring and comparing the spatial location accuracy by analyzing the Lee Sallee shape index (LSSI) of the damage areas. The findings confirm the correction of the HQPF based on machine learning and indicate its applicability to debris flow models.

scholarly journals The Development of a Quantitative Precipitation Forecast Correction Technique Based on Machine Learning for Hydrological Applications

Atmosphere ◽  
2020 ◽  
Vol 11 (1) ◽  
pp. 111 ◽  
Author(s):  
Chul-Min Ko ◽  
Yeong Yun Jeong ◽  
Young-Mi Lee ◽  
Byung-Sik Kim
Keyword(s):  
Machine Learning ◽  
Heavy Rainfall ◽  
Extreme Rainfall ◽  
Machine Learning Techniques ◽  
Precipitation Forecast ◽  
Machine Learning Technique ◽  
Rainfall Forecast ◽  
Quantitative Precipitation Forecast ◽  
Correction Technique ◽  
Learning Technique

This study aimed to enhance the accuracy of extreme rainfall forecast, using a machine learning technique for forecasting hydrological impact. In this study, machine learning with XGBoost technique was applied for correcting the quantitative precipitation forecast (QPF) provided by the Korea Meteorological Administration (KMA) to develop a hydrological quantitative precipitation forecast (HQPF) for flood inundation modeling. The performance of machine learning techniques for HQPF production was evaluated with a focus on two cases: one for heavy rainfall events in Seoul and the other for heavy rainfall accompanied by Typhoon Kong-rey (1825). This study calculated the well-known statistical metrics to compare the error derived from QPF-based rainfall and HQPF-based rainfall against the observational data from the four sites. For the heavy rainfall case in Seoul, the mean absolute errors (MAE) of the four sites, i.e., Nowon, Jungnang, Dobong, and Gangnam, were 18.6 mm/3 h, 19.4 mm/3 h, 48.7 mm/3 h, and 19.1 mm/3 h for QPF and 13.6 mm/3 h, 14.2 mm/3 h, 33.3 mm/3 h, and 12.0 mm/3 h for HQPF, respectively. These results clearly indicate that the machine learning technique is able to improve the forecasting performance for localized rainfall. In addition, the HQPF-based rainfall shows better performance in capturing the peak rainfall amount and spatial pattern. Therefore, it is considered that the HQPF can be helpful to improve the accuracy of intense rainfall forecast, which is subsequently beneficial for forecasting floods and their hydrological impacts.

scholarly journals Application of Machine Learning to Debris Flow Susceptibility Mapping along the China–Pakistan Karakoram Highway

2020 ◽  
Vol 12 (18) ◽  
pp. 2933
Author(s):  
Feng Qing ◽  
Yan Zhao ◽  
Xingmin Meng ◽  
Xiaojun Su ◽  
Tianjun Qi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Remote Sensing ◽  
Debris Flow ◽  
Debris Flows ◽  
Susceptibility Mapping ◽  
Support Vector ◽  
Safe Operation ◽  
Extreme Gradient Boosting ◽  
Karakoram Highway ◽  
Debris Flow Susceptibility

The China–Pakistan Karakoram Highway is an important land route from China to South Asia and the Middle East via Pakistan. Due to the extremely hazardous geological environment around the highway, landslides, debris flows, collapses, and subsidence are frequent. Among them, debris flows are one of the most serious geological hazards on the Karakoram Highway, and they often cause interruptions to traffic and casualties. Therefore, the development of debris flow susceptibility mapping along the highway can potentially facilitate its safe operation. In this study, we used remote sensing, GIS, and machine learning techniques to map debris flow susceptibility along the Karakoram Highway in areas where observation data are scarce and difficult to obtain by field survey. First, the distribution of 544 catchments which are prone to debris flow were identified through visual interpretation of remote sensing images. The factors influencing debris flow susceptibility were then analyzed, and a total of 17 parameters related to geomorphology, soil materials, and triggering conditions were selected. Model training was based on multiple common machine learning methods, including Ensemble Methods, Gaussian Processes, Generalized Linear models, Navies Bayes, Nearest Neighbors, Support Vector Machines, Trees, Discriminant Analysis, and eXtreme Gradient Boosting. Support Vector Classification (SVC) was chosen as the final model after evaluation; its accuracy (ACC) was 0.91, and the area under the ROC curve (AUC) was 0.96. Among the factors involved in SVC, the Melton Ratio (MR) was the most important, followed by drainage density (DD), Hypsometric Integral (HI), and average slope (AS), indicating that geomorphic conditions play an important role in predicting debris flow susceptibility in the study area. SVC was used to map debris flow susceptibility in the study area, and the results will potentially facilitate the safe operation of the highway.

scholarly journals Debris Flow Susceptibility Mapping Using Machine-Learning Techniques in Shigatse Area, China

2019 ◽  
Vol 11 (23) ◽  
pp. 2801 ◽  
Author(s):  
Yonghong Zhang ◽  
Taotao Ge ◽  
Wei Tian ◽  
Yuei-An Liou
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Debris Flow ◽  
Debris Flows ◽  
Gradient Boosting ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Triggering Factors ◽  
Extreme Gradient Boosting ◽  
Debris Flow Susceptibility

Debris flows have been always a serious problem in the mountain areas. Research on the assessment of debris flows susceptibility (DFS) is useful for preventing and mitigating debris flow risks. The main purpose of this work is to study the DFS in the Shigatse area of Tibet, by using machine learning methods, after assessing the main triggering factors of debris flows. Remote sensing and geographic information system (GIS) are used to obtain datasets of topography, vegetation, human activities and soil factors for local debris flows. The problem of debris flow susceptibility level imbalances in datasets is addressed by the Borderline-SMOTE method. Five machine learning methods, i.e., back propagation neural network (BPNN), one-dimensional convolutional neural network (1D-CNN), decision tree (DT), random forest (RF), and extreme gradient boosting (XGBoost) have been used to analyze and fit the relationship between debris flow triggering factors and occurrence, and to evaluate the weight of each triggering factor. The ANOVA and Tukey HSD tests have revealed that the XGBoost model exhibited the best mean accuracy (0.924) on ten-fold cross-validation and the performance was significantly better than that of the BPNN (0.871), DT (0.816), and RF (0.901). However, the performance of the XGBoost did not significantly differ from that of the 1D-CNN (0.914). This is also the first comparison experiment between XGBoost and 1D-CNN methods in the DFS study. The DFS maps have been verified by five evaluation methods: Precision, Recall, F1 score, Accuracy and area under the curve (AUC). Experiments show that the XGBoost has the best score, and the factors that have a greater impact on debris flows are aspect, annual average rainfall, profile curvature, and elevation.

Formation of dry granular fronts and watery tails indebris flows

2021 ◽  
Author(s):  
Xiannan Meng ◽  
Chris Johnson ◽  
Nico Gray
Keyword(s):  
Particle Size ◽  
Debris Flow ◽  
Debris Flows ◽  
Plug Flow ◽  
Shear Velocity ◽  
Formation Mechanisms ◽  
Size Segregation ◽  
Flume Experiments ◽  
Flow Models ◽  
Averaged Model

<p>Dry granular fronts and watery tails often develop in debris flows, but their formation mechanisms are still poorly understood. Dry bouldery debris flow fronts are often attributed to particle-size segregation, but idealized experimental mixtures of fluid and mono-disperse grains also exhibit the formation of dry fronts. This motivates the development of a new depth-averaged model that treats grain-water mixtures as a buoyancy and Darcy drag coupled multiphase medium. This system is able to describe the temporal and spatial evolution of the grain and water depths as well as the associated grain and water depth-averaged velocities. It considers the layered development of the flow and incorporates a shear velocity profile into the model, instead of the standard plug flow assumption that is employed by almost all debris-flow models. By revisiting Davies' moving bed flume experiments, it is shown that, in the under-saturated region, shear results in the surface layer of dry grains moving faster than the bulk and they are preferentially transported to the flow front to develop a dry snout. Conversely, in the over-saturated region, the flow thickness is sufficiently small that the water friction is stronger than the friction acting on the grains. As a result, the surface grains can move faster than the water and leave it behind. This novel theory provides a rational framework that describes the complete longitudinal profile of debris flows from the dry granular front to the pure watery tail without the need to consider particle-size segregation.</p>

scholarly journals Technical notes: Rainfall threshold calculation for debris flow early warning in areas with scarcity of data

2017 ◽  
Author(s):  
Hua-li Pan ◽  
Yuan-jun Jiang ◽  
Jun Wang ◽  
Guo-qiang Ou
Keyword(s):  
Debris Flow ◽  
Early Warning ◽  
Debris Flows ◽  
Empirical Relationship ◽  
Early Warning Systems ◽  
Rainfall Threshold ◽  
Negative Slope ◽  
Observation Data ◽  
Flow Forming ◽  
Initiation Mechanism

Abstract. Debris flows are one of the natural disasters that frequently occur in mountain areas, usually accompanied by serious loss of lives and properties. One of the most used approaches to mitigate the risk associated to debris flows is the implementation of early warning systems based on well calibrated rainfall thresholds. However, many mountainous areas have little data regarding rainfall and hazards, especially in debris flow forming regions. Therefore, the traditional statistical analysis method that determines the empirical relationship between rainfall and debris flow events cannot be effectively used to calculate reliable rainfall thre-shold in these areas. To solve this problem, this paper developed a quantitative method to identify rainfall threshold for debris flow early warning in data-poor areas based on the initiation mechanism of hydraulic-driven debris flow. First, we studied the characteristics of the study area, including meteorology, hydrology, topography and physical characteristics of the loose solid materials. Then, the rainfall threshold was calculated by the initiation me-chanism of the hydraulic debris flow. The results show that the proposed rainfall threshold curve is a function of the antecedent precipitation index and 1-h rainfall. The function is a line with a negative slope. To test the proposed method, we selected the Guojuanyan gully, a typical debris flow valley that during the 2008–2013 period experienced several debris flow events and that is located in the meizoseismal areas of Wenchuan earthquake, as a case study. We compared the calculated threshold with observation data, showing that the accuracy of the method is satisfying and thus can be used for debris flow early warning in areas with scaricty of data.

scholarly journals Modeling the Spatial Distribution of Debris Flows and Analysis of the Controlling Factors: A Machine Learning Approach

2021 ◽  
Vol 13 (23) ◽  
pp. 4813
Author(s):  
Yan Zhao ◽  
Xingmin Meng ◽  
Tianjun Qi ◽  
Guan Chen ◽  
Yajun Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Spatial Distribution ◽  
Debris Flow ◽  
Debris Flows ◽  
Controlling Factors ◽  
Central China ◽  
Mitigation Measures ◽  
Ground Acceleration ◽  
Total Contribution ◽  
Mountainous Regions

Debris flows are a major geological hazard in mountainous regions. For improving mitigation, it is important to study the spatial distribution and factors controlling debris flows. In the Bailong River Basin, central China, landslides and debris flows are very well developed due to the large differences in terrain, the complex geological environment, and concentrated rainfall. For analysis, 52 influencing factors, statistical, machine learning, remote sensing and GIS methods were used to analyze the spatial distribution and controlling factors of 652 debris flow catchments with different frequencies. The spatial distribution of these catchments was divided into three zones according to their differences in debris flow frequencies. A comprehensive analysis of the relationship between various factors and debris flows was made. Through parameter optimization and feature selection, the Extra Trees classifier performed the best, with an accuracy of 95.6%. The results show that lithology was the most important factor controlling debris flows in the study area (with a contribution of 26%), followed by landslide density and factors affecting slope stability (road density, fault density and peak ground acceleration, with a total contribution of 30%). The average annual frequency of daily rainfall > 20 mm was the most important triggering factor (with a contribution of 7%). Forest area and vegetation cover were also important controlling factors (with a total contribution of 9%), and they should be regarded as an important component of debris flow mitigation measures. The results are helpful to improve the understanding of factors influencing debris flows and provide a reference for the formulation of mitigation measures.

Socially Shared Regulation of Learning: A Review

2015 ◽  
Vol 20 (3) ◽  
pp. 190-203 ◽  
Author(s):  
Ernesto Panadero ◽  
Sanna Järvelä
Keyword(s):  
Empirical Evidence ◽  
Learning Theory ◽  
Qualitative Data ◽  
Social Regulation ◽  
Observation Data ◽  
Optimal Form ◽  
Self Regulated Learning ◽  
The Past ◽  
Recorded Observation ◽  
Socially Shared Regulation

Abstract. Socially shared regulation of learning (SSRL) has been recognized as a new and growing field in the framework of self-regulated learning theory in the past decade. In the present review, we examine the empirical evidence to support such a phenomenon. A total of 17 articles addressing SSRL were identified, 13 of which presented empirical evidence. Through a narrative review it could be concluded that there is enough data to maintain the existence of SSRL in comparison to other social regulation (e.g., co-regulation). It was found that most of the SSRL research has focused on characterizing phenomena through the use of mixed methods through qualitative data, mostly video-recorded observation data. Also, SSRL seems to contribute to students’ performance. Finally, the article discusses the need for the field to move forward, exploring the best conditions to promote SSRL, clarifying whether SSRL is always the optimal form of collaboration, and identifying more aspects of groups’ characteristics.

Recent Progress in Machine Learning-based Prediction of Peptide Activity for Drug Discovery

2019 ◽  
Vol 19 (1) ◽  
pp. 4-16 ◽  
Author(s):  
Qihui Wu ◽  
Hanzhong Ke ◽  
Dongli Li ◽  
Qi Wang ◽  
Jiansong Fang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Drug Discovery ◽  
Large Scale ◽  
Recent Progress ◽  
High Specificity ◽  
Learning Approaches ◽  
Anticancer Peptides ◽  
The Past ◽  
Traditional Approaches ◽  
Large Scale Screening

Over the past decades, peptide as a therapeutic candidate has received increasing attention in drug discovery, especially for antimicrobial peptides (AMPs), anticancer peptides (ACPs) and antiinflammatory peptides (AIPs). It is considered that the peptides can regulate various complex diseases which are previously untouchable. In recent years, the critical problem of antimicrobial resistance drives the pharmaceutical industry to look for new therapeutic agents. Compared to organic small drugs, peptide- based therapy exhibits high specificity and minimal toxicity. Thus, peptides are widely recruited in the design and discovery of new potent drugs. Currently, large-scale screening of peptide activity with traditional approaches is costly, time-consuming and labor-intensive. Hence, in silico methods, mainly machine learning approaches, for their accuracy and effectiveness, have been introduced to predict the peptide activity. In this review, we document the recent progress in machine learning-based prediction of peptides which will be of great benefit to the discovery of potential active AMPs, ACPs and AIPs.

Sign in / Sign up

Export Citation Format

Share Document