Recent Advances of Deep Learning in Geological Hazard Forecasting

Geological hazard is an adverse geological condition that can cause loss of life and property.Accurate prediction and analysis of geological hazards is an important and challenging task.In the past decade,there has been a great expansion of geohazard detection data and advancement in data-driven simulation techniques.In particular,great efforts have been made in applying deep learning to predict geohazards.To understand the recent progress in this field,this paper provides an overview of the commonly used data sources and deep neural networks in the prediction of a variety of geological hazards.

Thinking on Effective Methods of Successful Prediction of Geological Hazards in Shaanxi, China

In the process of human survival and development, it is inevitable to transform the original state of the world, thus forming a contradiction between the earth and the earth. The violent form of this contradiction is geological disasters. Geological disasters pose a threat to human life and property, and cause damage caused by natural or human factors, often causing casualties. The destruction process of geological disasters is usually a gradual process, showing many pre-disaster symptoms, such as local landslides, surface cracks, building deformation, tree skew, and ground sound. Evacuation can be avoided in advance according to the disaster precursors, so as to avoid casualties and achieve successful prediction. By reviewing the general situation of geological disasters in Shaanxi Province and the casualties in 2020, the difficulties in the prevention and control of geological disasters are summarized. In view of these difficulties, an on-site investigation, visit and analysis of geological disaster points and successful forecast points in Shaanxi Province in 2020 were conducted. In addition, combined with actual cases and years of work experience, the successful prediction experience of geological disasters was discussed from 8 aspects. Finally, the “Regulations on the Reward for Successful Geological Disaster Forecasting in Shaanxi Province” was revised in order to improve the successful prediction ability of geological disasters in Shaanxi Province and even the whole country, provide reference for future prevention and control of geological disasters, and effectively protect the safety of people’s lives and property.

Predicting geological hazards during tunnel construction

The complicated geological conditions and geological hazards are challenging problems during tunnel construction,which will cause great losses of life and property.Therefore,reliable prediction of geological defective features,such as faults,karst caves and groundwater,has important practical significances and theoretical values.In this paper,we presented the criteria for detecting typical geological anomalies using the tunnel seismic prediction（TSP） method.The ground penetrating radar（GPR） signal response to water-bearing structures was used for theoretical derivations.And the 3D tomography of the transient electromagnetic method（TEM） was used to develop an equivalent conductance method.Based on the improvement of a single prediction technique,we developed a technical system for reliable prediction of geological defective features by analyzing the advantages and disadvantages of all prediction methods.The procedure of the application of this system was introduced in detail.For prediction,the selection of prediction methods is an important and challenging work.The analytic hierarchy process（AHP） was developed for prediction optimization.We applied the newly developed prediction system to several important projects in China,including Hurongxi highway,Jinping II hydropower station,and Kiaochow Bay subsea tunnel.The case studies show that the geological defective features can be successfully detected with good precision and efficiency,and the prediction system is proved to be an effective means to minimize the risks of geological hazards during tunnel construction.

Co-seismic Faults and Geological Hazards and Incidence of Active Fault of Wenchuan Ms 8.0 Earthquake,Sichuan,China

There are two co-seismic faults which developed when the Wenchuan earthquake happened. One occurred along the active fault zone in the central Longmen Mts. and the other in the front of Longmen Mts. The length of which is more than 270 kin and about 80 km respectively. The co-seismic fault shows a reverse flexure belt with strike of N45°-60°E in the ground, which caused uplift at its northwest side and subsidence at the southeast. The fault face dips to the northwest with a dip angle ranging from 50° to 60°. The vertical offset of the co-seismic fault ranges 2.5-3.0 m along the Yingxiu- Beichuan co-seismic fault, and 1.5-1.1 m along the Doujiangyan-Hanwang fault. Movement of the coseismic fault presents obvious segmented features along the active fault zone in central Longmen Mts. For instance, in the section from Yingxiu to Leigu town, thrust without evident slip occurred; while from Beichuan to Qingchuan, thrust and dextral strike-slip take place. Main movement along the front Longmen Mts. shows thrust without slip and segmented features. The area of earthquake intensity more than IX degree and the distribution of secondary geological hazards occurred along the hanging wall of co-seismic faults, and were consistent with the area of aftershock, and its width is less than 40km from co-seismic faults in the hanging wall. The secondary geological hazards, collapses, landslides, debris flows et al., concentrated in the hanging wall of co-seismic fault within 0-20 km from co-seismic fault.

Distribution Characteristics and Influencing Factors of Geological Hazards in Tibet

Tibet is one of the areas with most serious geological hazards in China, and the distribution of disasters has obvious local charac teristics. Tibet can be classified as three parts through zoning the danger degree, the mountain canyon high danger zone of east and southeast Tibet, the plateau mountain lake basin and valley middle danger zone of south Tibet, and the Plateau Mountain lake basin low danger zone of south Tibet. This paper takes the debris flow, collapse, landslide as the key points to analyze the distribution characteristics of geological hazards, and analyze the factors which influence the distribution of geological hazards, such as terrain landform, formation lithology, geologic structure pattern, precipitation, earthquake, human activity and so on. finally, as a conclusion., in whole Tibet, the geological hazards are more in southeast than in northwest, more in mountainous area which in the edge of plateau and river valley than in the interior of plateau and lake basin. And most hazards distribute in the regions where human activity is stronger than in other regions, for example towns or strips along the highway.

Integrated Evaluation on Highway Engineering Geological Hazard in Mountainous Area of Enshi, Hubei

Aiming at the geological features of highway engineering in mountainous area of Enshi, Hubei Province, the principles to set up an integrated evaluation system for highway engineering geological hazard are formulated. Then, the integrated evaluation system for highway engineering geological hazard in mountainous area of Enshi is established. In the evaluation system, the first-level evaluation indices are geological development degree, geological conditions, and damaging power, whereas the second-level indices including key factors affecting main kinds of geological hazard. Meanwhile, the borders of indices are determined. At last, the method of Fuzzy Comprehensive Evaluation （FCE） is adopted to quantitatively evaluate the highway engineering geological hazard in mountainous area of Enshi.

Regional Integrated Meteorological Forecasting and Warning Model for Geological Hazards Based on Logistic Regression

Information model is adopted to integrate factors of various geosciences to estimate the susceptibility of geological hazards. Further combining the dynamic rainfall observations, Logistic regression is used for modeling the probabilities of geological hazard occurrences, upon which hierarchical warnings for rainfall-induced geological hazards are produced. The forecasting and warning model takes numerical precipitation forecasts on grid points as its dynamic input, forecasts the probabilities of geological hazard occurrences on the same grid, and translates the results into likelihoods in the form of a 5-level hierarchy. Validation of the model with observational data for the year 2004 shows that 80% of the geological hazards of the year have been identified as ＂likely enough to release warning messages＂. The model can satisfy the requirements of an operational warning system, thus is an effective way to improve the meteorological warnings for geological hazards.

GIS-Based Highway Geological Hazard Information Management and Decision Support System

Based on the practical application of Geology Information System（GIS） throughout the world, combined with the characters of road＇s geological hazard and it＇s supervision, the paper introduces on the importance of the research on road＇s geological hazards information management and decision-making support system. The paper also analyzes the system＇s target, the principles and key techniques in developing the system. In the research, we developed the GIS-based road＇s geological hazard information management and decision-making support system and applied it to one speedway in the west of China where contains typical geological hazards. The system based on the database of road＇s geological hazard on the grounds of spatial graphic information and attribute information. By virtue of the scientific assessment and prediction mathematical model, integrating the GIS＇s strongpoint on spatial analyzing, the system is capable of visualizing the regionalization of road according to the geological hazards it contains, and accurately assessing and predicting geological hazards, thus efficiently assists the road construction and management units in the decision making on controlling the geological hazards and reducing the related loss.

Methodological study of coastal geological hazard assessment based on GIS

The current researches on risk assessment of geological disasters mainly focus on unexpected disasters such as collapses, landslides and mud-rock flows etc. As the convergence zone of land and sea, coastal zone is the most active and complex area of interactions of lithosphere, hydrosphere, atmosphere, biosphere and anthroposphere. The ecological environment of coastal zone is very fragile, so further systematical research on coastal geological hazard assessment and prevention is in urgent need. The author begins with the definition and research contents and selects three typical coastal geological disasters, namely, the seawater intrusion, coastline change and sea-level rise as the objects of study. The systematic analysis and study on assessment system and methods are conducted, hazard assessment factors are selected, and a completely set of coastal disaster assessment system is established based on the technique of GIS. We took Bao’an District of Shenzhen City as an example and carried out a case study.

Types and Distribution of Geological Hazards in the South China Sea

Various types of geological hazards exist in the South China Sea. In dynamics sense, they can be categorized into 5 principal genetic types related to effects of hydraulic dynamics, gaseous activity, soil mechanics, gravity and tectonism, respectively. Integrated analyses indicate that the geological hazards associated with volcanoes, earthquakes and fractures are mainly distributed in tectonically active regions, whereas those resulting from mudflows, landslides and diapirs are usually concentrated in the region of slope, that shallow gas, high pressure gas pockets and soft intercalations are major potential geological hazards in the inner shelf, and that strong hydraulic dynamics, especially storm tide, is one of the major causes of geological hazards in the littoral areas. The geological hazards that occurred in the South China Sea are also characterized by periodicity, succession and, to a certain extent, unpredictability in addition to regionalization.

Chinese public participation monitoring and warning system for geological hazards

In China,many geological hazards occurred in remote mountainous regions,and it was time-consuming to disseminate disaster information for the responsible parties to make timely judgements.Besides,only relying on professionals to monitor and manage disasters was demanding and costly.The Chinese government created a system to engage residents in the process of disaster risk management,namely’Public Participation Monitoring and Warning’(PPMW),to disseminate timely disaster information and bring down management costs.The objective of this system was to reduce casualties with minimum cost by organizing residents to evacuate from disasters in advance.This paper introduced the PPMW system,including its structure,operation mechanism by reviewing government documents and research articles,and its implementation by a case study of a landslide at Boli village(E 101°01’,N 27°29’),Yanyuan County,Sichuan Province,China on July 19th 2018.Further,this paper analyzed the strengths and limitations of the system and discussed its future development.It had the potential to become an affordable disaster risk management tool for other countries facing similar situations to China.

Geological Hazards Occurred on the Road Connecting Vashlijvari-Lisi (M. Machavariani Street) in Tbilisi

The capital of Georgia—Tbilisi has a very convenient location and is a node of the transit corridor. Along with natural-geological conditions, its complexity is due to the rapid demographic growth of the city in a highly “sensitive” area of the geological environment and the pressure of high engineering and agricultural activities. In Tbilisi, it is observed almost all types of geological hazards, including landslide-gravitational, suffosion, debris/mudflows, river bank erosion and inundation were caused by groundwater. These hazards cause high damages to the residential houses and other infrastructure facilities. Most importantly and most tragically is that these kind of negative geological events are often accompanied by human casualties. The study discusses the geological processes developed in March 2021 in the corridor of the Vashlijvari-Lisi road (Machavariani Street). The information obtained from the study, reflects the triggering factors of the geological hazards, also damages caused by them, and provides recommendations for short-term and long-term protective measures that should ensure the sustainable operation of the road and other infrastructure facilities.

Application of oblique photogrammetry technique in geological hazard identification and decision management

With the continuous development of the oblique photography technique, it has been used more and more widely in the field of geological disasters. It can quickly obtain the three-dimensional(3D) real scene model of dangerous mountainous areas under the premise of ensuring the safety of personnel while restoring the real geographic information as much as possible. However, geological disaster areas are often accompanied by many adverse factors such as cliffs and dense vegetation. Based on this, the paper introduced the flight line design of oblique photogrammetry, analyzed the multi-platform data fusion processing, studied the multi-period data dynamic evaluation technology and proposed the application methods of data acquisition, early warning, disaster assessment and decision management suitable for geological disaster identification through the analysis of actual cases, which will help geologists to plan and control geological work more scientifically and rationally, improve work efficiency and reduce the potential personnel safety hazards in the process of geological survey, to offer technical support to the application of oblique photogrammetry in geological disaster identification and decision making and provide the scientific basis for personal and property safety protection and later-stage geological disaster management in disaster areas.

Integrating NLP and Ontology Matching into a Unified System for Automated Information Extraction from Geological Hazard Reports

Many detailed data on past geological hazard events are buried in geological hazard reports and have not been fully utilized. The growing developments in geographic information retrieval and temporal information retrieval offer opportunities to analyse this wealth of data to mine the spatiotemporal evolution of geological disaster occurrence and enhance risk decision making. This study presents a combined NLP and ontology matching information extraction framework for automatically recognizing semantic and spatiotemporal information from geological hazard reports. This framework mainly extracts unstructured information from geological disaster reports through named entity recognition, ontology matching and gazetteer matching to identify and annotate elements, thus enabling users to quickly obtain key information and understand the general content of disaster reports. In addition, we present the final results obtained from the experiments through a reasonable visualization and analyse the visual results. The extraction and retrieval of semantic information related to the dynamics of geohazard events are performed from both natural and human perspectives to provide information on the progress of events.

Deformation,structure and potential hazard of a landslide based on InSAR in Banbar county,Xizang(Tibet)

The Tibetan Plateau is characterized by complex geological conditions and a relatively fragile ecological environment.In recent years,there has been continuous development and increased human activity in the Tibetan Plateau region,leading to a rising risk of landslides.The landslide in Banbar County,Xizang(Tibet),have been perturbed by ongoing disturbances from human engineering activities,making it susceptible to instability and displaying distinct features.In this study,small baseline subset synthetic aperture radar interferometry(SBAS-InSAR)technology is used to obtain the Line of Sight(LOS)deformation velocity field in the study area,and then the slope-orientation deformation field of the landslide is obtained according to the spatial geometric relationship between the satellite’s LOS direction and the landslide.Subsequently,the landslide thickness is inverted by applying the mass conservation criterion.The results show that the movement area of the landslide is about 6.57×10^(4)m^(2),and the landslide volume is about 1.45×10^(6)m^(3).The maximum estimated thickness and average thickness of the landslide are 39 m and 22 m,respectively.The thickness estimation results align with the findings from on-site investigation,indicating the applicability of this method to large-scale earth slides.The deformation rate of the landslide exhibits a notable correlation with temperature variations,with rainfall playing a supportive role in the deformation process and displaying a certain lag.Human activities exert the most substantial influence on the spatial heterogeneity of landslide deformation,leading to the direct impact of several prominent deformation areas due to human interventions.Simultaneously,utilizing the long short-term memory(LSTM)model to predict landslide displacement,and the forecast results demonstrate the effectiveness of the LSTM model in predicting landslides that are in a continuous development and movement phase.The landslide is still active,and based on the spatial heterogeneity of landslide deformation,new recommendations have been proposed for the future management of the landslide in order to mitigate potential hazards associated with landslide instability.

Automated machine learning for rainfall-induced landslide hazard mapping in Luhe County of Guangdong Province,China

Landslide hazard mapping is essential for regional landslide hazard management.The main objective of this study is to construct a rainfall-induced landslide hazard map of Luhe County,China based on an automated machine learning framework(AutoGluon).A total of 2241 landslides were identified from satellite images before and after the rainfall event,and 10 impact factors including elevation,slope,aspect,normalized difference vegetation index(NDVI),topographic wetness index(TWI),lithology,land cover,distance to roads,distance to rivers,and rainfall were selected as indicators.The WeightedEnsemble model,which is an ensemble of 13 basic machine learning models weighted together,was used to output the landslide hazard assessment results.The results indicate that landslides mainly occurred in the central part of the study area,especially in Hetian and Shanghu.Totally 102.44 s were spent to train all the models,and the ensemble model WeightedEnsemble has an Area Under the Curve(AUC)value of92.36%in the test set.In addition,14.95%of the study area was determined to be at very high hazard,with a landslide density of 12.02 per square kilometer.This study serves as a significant reference for the prevention and mitigation of geological hazards and land use planning in Luhe County.

Analysis of debris flow control effect and hazard assessment in Xinqiao Gully,Wenchuan M_(s)8.0 earthquake area based on numerical simulation

Xinqiao Gully is located in the area of the 2008 Wenchuan M_(s)8.0 earthquake in Sichuan province,China.Based on the investigation of the 2023"6-26"Xinqiao Gully debris flow event,this study assessed the effectiveness of the debris flow control project and evaluated the debris flow hazards.Through field investigation and numerical simulation methods,the indicators of flow intensity reduction rate and storage capacity fullness were proposed to quantify the effectiveness of the engineering measures in the debris flow event.The simulation results show that the debris flow control project reduced the flow intensity by41.05%to 64.61%.The storage capacity of the dam decreases gradually from upstream to the mouth of the gully,thus effectively intercepting and controlling the debris flow.By evaluating the debris flow of different recurrence intervals,further measures are recommended for managing debris flow events.

Machine Learning-Based Evaluation of Susceptibility to Geological Hazards in the Hengduan Mountains Region, China

The Hengduan Mountains Region(HMR) is one of the areas that experience the most frequent geological hazards in China. However, few reports are available that address the geological hazard susceptibility of the region.This study developed six machine learning models to assess the geological hazard susceptibility. The results show that areas with medium and high susceptibility to geological hazards as a whole account for almost 21% of the total area, while both are 18% when it comes to the single hazard of landslide and rockfall respectively. Medium and high geological hazard susceptibility is found in three parts of the HMR with different characteristics:(1)the central and southern parts, where the population of the region concentrates;(2) the northern part, where higher geological hazard susceptibility is found along the mountain ranges;and(3) the junction of Tibet, Yunnan, and Sichuan in the eastern part, which is prone to larger-scale geological hazards. Of all the potential influencing factors,topographic features and climatic variables act as the major driving factors behind geological hazards and elevation,slope, and precipitation are crucial indicators for geological hazard susceptibility assessment. This study developed the geological hazard susceptibility maps of the HMR and provided information for the multi-hazard risk assessment and management of the region.

Application of remote sensing for investigating mining geological hazards

To investigate geological mining hazards using digital techniques such as highresolution remote sensing,a semi-automatically geological mining hazards extraction method is proposed based on the case of the Shijiaying coal mine,located in Fangshan District,Beijing,China.In the method,the vegetation is first removed using the normalized difference vegetation index(NDVI)on the GeoEye-1 data.Then,geological mining hazards interpretation features are determined after color enhancement using principal component analysis(PCA)transformation.Bitmaps mainly covered by geological mining hazards are isolated by masking operation in the environment for visualizing images software.Next,each bitmap is classified into a two-valued imagery using support vector machine algorithm.In the two-valued imagery,1 denotes the geological mining hazards,while 0 denotes none.Afterwards,the two-valued imagery is converted into a vector graph by corresponding functions in the ArcGIS software and no geological mining hazards regions in the vector graph are deleted manually.Finally,the correlation between factors(such as mining activity,lithology,geological structure,and slope)and geological mining hazards is analyzed using a logistic regression and a hazardous-area forecasting model is built.The results of field verification show that the accuracy of the geological mining hazards extraction method is 98.1%and the results of the hazardous-area forecasting indicate that the logistic regression is an effective model in assessing geological hazard risks and that mining activity is the main contributing factor to the hazards,while geological structure,slope,lithology,roughness of the surface,and aspect are the secondary.

GEOLOGICAL HAZARD CONTROL IN CHINA

ＧＥＯＬＯＧＩＣＡＬＨＡＺＡＲＤＣＯＮＴＲＯＬＩＮＣＨＩＮＡＺｈａｎｇＨｏｎｇｒｅｎ（张宏仁）（Ｖｉｃｅ－ＭｉｎｉｓｔｅｒｏｆｔｈｅＭｉｎｉｓｔｒｙｏｆＧｅｏｌｏｇｙａｎｄＭｉｎｅｒａｌＲｅｓｏｕｒｃｅｓ，ｔｈｅＰＲＣ）Ａｍｏｎｇｔｈｅａｒｄｕｏｕｓ...