Currently, the technology of large language models is evolving rapidly and accelerating its integration in geological disaster prevention and control. It has been expanding the application scenarios and breaking the limitations in data analysis and complex modeling capabilities as well as innovating the traditional research paradigm. To further promote new breakthroughs in AI technologies in the intelligent prevention and control of geological disasters, this article reviews the evolution characteristics of large language model technology and the application scenarios in multiple fields, and also discusses the key technologies including small sample learning, multimodal data fusion, lightweight model and transfer application, as well as expert knowledge embedding and human-computer collaboration, which are also the main ideas and research focus directions for achieving intelligent identification of geological disaster hazards. The article also proposes an "AI + geological disasters" research framework, technical ideas and typical application scenarios based on core elements including "application scenarios, key issues, mechanism of action, data modalities, sample characteristics, model development, expert knowledge, and human-computer collaboration". This highlights the important value of AI technology in geological disasters research in solving the dealing with multi-dimensional, multi-scale, nonlinear and complex relationship modeling problems. The purpose of this article is to promote AI technologies to integrate into geological disaster prevention and control work at a deeper level, from data, models, and knowledge, and also better leverage AI technology to promote the development of disaster prevention and mitigation towards a greater precision and intelligence.

This paper presents a preliminary investigation of the instability and landslide of a road subgrade slope that occurred on May 1, 2024 along Mei–Da Expressway. The investigation method is on the basis of past investigation results and findings about instability and landslide of road subgrade slopes in Hong Kong. The Mei–Da Expressway subgrade was constructed by cutting and filling on mountainous slopes. The landslide occurred at a fill slope along a gully. The fill soil was excavated from local soil of completely decomposed granite (CDG). The paper tests the particle size distribution and liquid and plastic limits of the three CDG soil samples that were excavated from adjacent natural slopes. It then examines the stability of the fill slope with a simplified geological slope model. It further investigates the catchments before and after the road construction. It then presents a possible mechanism for the slope instability and landslide. The field observations by others can confirm this mechanism. The cutting and filling of the hillside slopes significantly changed the runoff paths of the three natural catchments above the road. The road can intercept the runoff along the three gullies and direct all the runoff into the fill slope. The large volume of rainfall water can erode and liquefy the fills, which can cause the instability and landslide of the road subgrade slope. There is an urgent need to further investigate this mode of road subgrade instability and landslide since it is not uncommon.

The geological hazards of submarine landslides can cause serious damage to infrastructure such as offshore wind power, submarine optical cables, and marine platforms, posing a serious challenge to the major strategic task of building a maritime power and ensuring the geological safety of marine engineering. The article systematically reviews the research process of submarine landslide turbidity current geological hazards, summarizes the dynamic characteristics of submarine landslide-turbidity flow chain, dynamic erosion types, mechanisms of triggering, evolution, migration, erosion and sedimentation, theoretical models of erosion, and the influence of complex landforms such as uplift, canyons, and basins. A novel dynamic erosion approach is put forward of submarine landslide-turbidity flow chain, including quantitative, multiphase, whole process, erosion flow-state transformation. Finally, in view of the development of major projects such as offshore wind power, marine resource development, marine transportation, and marine engineering equipment, the geological model and identification technology are discussed of the erosion-prone structure of submarine landslide landslide-turbidity flow chain, as well as the composite, overlapping, and heterogeneous dynamic erosion mechanic model of the disaster chain, and the issues of prevention and control of boundary layer dynamic erosion.

Sedongpu gully, situated in the Yarlung Zangbo Grand Canyon in Xizang, is an area of frequent massive river-damming landslides that threaten the geological safety of border towns and major engineering projects. This study focused on a detailed analysis of two river-damming events that occurred on April 15 and May 14, 2024. The disaster formation processes, main causes, and developing trends were widely analyzed. From the methods of water level monitoring, ground motion monitoring, investigating by helicopter, and survey by high-altitude and Unmanned Aerial Vehicles, the formation and development processes of these river-damming events were identified and analyzed. It was found that the disaster body slid along the gully reached 8 minutes and the river-damming lasted over 10 hours. The second disaster was relatively more serious because the landslide-dammed lake had not completely burst, which significantly aggravated the damming of the main channel of the Yarlung Zangbo River. Their causes were examined from topographical and geological conditions, seismic factors, and climate factors. It was determined that the Sedongpu gully has large height differences, fractured rock structures, and a rich source of loose deposits, which provides favorable conditions for disaster formation; the temperature rising during the alternation of the spring and summer leads to the acceleration of glacier melting and the enhancement of hydrodynamic effect, triggering the occurrence of river-damming disaster chain. Furthermore, it is believed that Sedongpu gully has entered an active period of river-damming disaster chain, based on the interpretation of the comprehensive remote sensing images. Our findings revealed that the major river-damming hazards that occurred in 2018 had caused significant changes to the river morphology of the Yarlung Zangbo River; subsequent large-scale river-damming events resulted in more and more clogged up with the river channel, which increased the risk of forming a giant landslide dam. Finally, this paper provided some suggestions for addressing the issues related to river damming, rising water, outburst flooding, monitoring, early warning, and disaster reduction measures for the high-altitude and long runout disasters in the Sedongpu gully.

Geological disasters of ground collapse in mining area have a significant impact on infrastructure, such as roads, pipelines, residents’ lives and property, and safety production in mining area. Yingcheng gypsum mine in Hubei Province has a mining history of nearly 400 years, and long-term underground mining has formed a large range of goaf and ground collapse. Based on the systematic collection of pre-mining data and supplementary survey, the types and distribution rules of ground collapse were analyzed by multi-factor comprehensive and geological method. Based on the theory of "three zones", the cause mechanism of ground collapse of old hole type and mining cavity type were studied. The research shows that the ground collapse of gypsum mine is mainly small to medium in Yingcheng, and the geological hazards of ground collapse are divided into mining cavity type and old hole type. The caving type of ground collapse mainly includes two types: pillar breakage type and bending settlement type. Pillar breakage type is mainly the pillar and roof collapse caused by room and pillar mining, and bending settlement type is mainly the roof collapse caused by insufficient filling rate of longwall filling method mining. The main controlling factors of goaf-type surface collapse are the goaf filling condition and the ratio of depth to thickness. When the ratio of depth to thickness is less than 60, the surface collapse occurs more often, and with the increase of the ratio of depth to thickness, the ground collapse gradually decreases. The degree of subsidence deformation in the old hole type depends on whether the old hole is connected with the large-scale goaf and whether it is filled with water. The research results have guiding significance for gypsum mine risk management, safety assessment, monitoring and early warning system construction.

Following the catastrophic “5•12” Wenchuan earthquake, extensive debris was deposited on mountain surfaces in the earthquake zone, and significant vegetation damage occurred, providing abundant material for debris flow outbreaks and substantially increasing their risk. Previous studies primarily focused on vegetation recovery when assessing post-earthquake debris flow risks. However, field surveys revealed that large quantities of dynamic storage materials in the gullies significantly impact risk assessments. Based on field survey data, this study uses Qipan gully in Wenchuan County as a research subject and employs multi-source and multi-scale monitoring tools (Landsat series, Quick-bird, and UAVs) to analyze and statistically assess the source materials on slopes and gullies both pre- and post-earthquake. A dynamic risk assessment model for debris flow is constructed using game theory combined with a cloud model, assessing the risk from 2005 to 2019. Findings indicate that post-earthquake slope material sources were 7.7 times those pre-earthquake, and by 2019, with recovery to pre-earthquake levels by 2019. Statistical estimations based on recorded debris flow eruptions and sediment removal volumes show a reduction of approximately 7.813×10⁶ m³ in dynamic material sources by 2019. Assessing both slope and gully material sources yields more realistic results than considering slope sources alone. These results provide references and guidance for dynamic risk assessments of debris flow, impacting major engineering projects in increasingly seismic regions and effectively ensuring the safety of life and property.

Dammed lakes are characterized by suddenness,high impact and wide range of influence, which seriously threaten the lives and properties of people downstream. Rapid assessment of the potential loss of life is of great significance for emergency decision-making, emergency evacuation and other emergency response activities. In order to achieve timely, rapid and accurate assessment, this paper constructs a system of rapid assessment methods for barrier lake breaching loss of life based on simplified parameters, analyzing and calculating four modules respectively, barrier body stability, outburst flood peak, failure flood evolution and loss of life, and validating them using the Baige Dammed Lakes event with an accuracy rate of 65%. On this basis, the Shiping landslide, which has a high risk of river blockage in the Minjiang River Basin, was selected for prediction and analysis. The stability of the barrier body was assessed through the methodological system proposed in this paper, the peak flood flow under full breach conditions and its flood evolution were calculated, the impact range caused by the flood risk on the Wenchuan County town of Weizhou was delineated, and the possible missing or dead population caused by the flood was predicted. The results of the case calculations show that under partial alarm conditions, the Shiping landslide barrier lake breaching will result in a loss of population of approximately 753 people in Weizhou town, and that disaster prevention, mitigation and preparedness work such as monitoring and early warning, emergency planning and emergency evacuation drills need to be strengthened to reduce disaster risk. A basis for disaster prevention and mitigation planning and emergency response and decision-making for Dammed Lakes can be provided through the establishment of this rapid assessment method system for loss of life.

The prevention and response of geological hazards in linear engineering projects must consider safety throughout the project’s lifecycle. This involves specific strategies tailored to three distinct stages: planning and route selection, design and construction, and project operation. The planning and selection stage focuses on avoiding geological disaster risks. In the design and construction stage, engineering measures are implemented to prevent geological hazards from occurring or triggering. During project operation stage, the emphasis shift to monitoring, early warning, and efficient management of ongoing or potential geological hazards.

The mound landslides in the Qinba Mountain area are characterized by their large number, wide distribution, high density, and high frequency occurence, with excavation-induced landslides being particularly severe. This study focuses on the gravel soils from the mine hole landslide in the third group of Lingfeng Village, Xiaoling Town, Qinba Mountain area. Through large-scale direct shear tests conducted in the laboratory, an in-depth investigation of the gravelly soils in the study area was carried out. The variation rules of shear strength under different moisture contents, dry densities, and normal stresses were explored. Based on the results of these indoor tests, the Midas GTS NT finite element numerical simulation software was used to calculate and analyze the slope conditions before the occurrence of the landslide. Simulations were conducted to analyze the changes in stress, displacement, and stability of the slope after excavation and under two working conditions: post-excavation and post-excavation coupled with rainfall. Finally, the mechanism of typical excavation-induced landslides in stockpiles is summarized based on these findings. It was found by numerical simulation: human engineering activities, i.e., excavation of the toe of slopes and heavy rainfall occurring in the area, are the main triggering factors for landslides: the deformation pattern of typical excavation-induced landslides in the Qinba Mountains can be summarised as: traction-creep-slip type.The research results can provide valuable reference for the prevention and control of excavation-induced landslides in accumulation layers in the Qinba Mountain area.

The Nyingchi area exhibits complex topography, high fault activity, an extensive water systems distribution, variable climatic conditions, and frequent geological hazards. These factors have a significantly growing impact on the economic development and engineering construction in the entire Nyingchi area. Among these hazards, landslides, collapses, and debris flows are the most common geological hazards in Nyingchi area. In order to quantitatively analyze the sensitivity of hazards in the Nyingchi area to the impact factors, this study, based on GIS and the certainty coefficient analysis method, selected ten factors, including elevation, slope aspect, topographic relief, and topographic humidity index, to conduct sensitivity analysis on landslide, rockfall and debris flow hazards. The analysis results show that: (1) The sensitive range of factors influencing landslide, rockfall and debris flow hazards in the Nyingchi area include elevation between 0.82 and 3.79 km; slope aspects facing eastward, northeastward, southward, and westward; topographic relief ranging from 0 to 24 m/km²; distances from the water system within 0 to 3 km; normalized vegetation index ranging from 0.47 to 0.81; distances from the road wihtin 0 to 1.5 km; distances from the active fault zone within 0 to 3 km; annual average rainfall ranging from 51.15 to 146.14 mm; annual average temperatures between 4.02 and 17.22 °C. There exists a strong correlation between hazards and these impact factors. (2) Sensitivity among influencing factors follows this order: annual average temperature > distance from water system > elevation > topographic relief > distance from road > normalized vegetation index > average annual rainfall > topographic humidity index > distance from the active fault > aspect. The research results provide references for engineering construction and hazards prevention and mitigation work in the Nyingchi area.

With global climate warming, the rate of snow and ice melting has accelerated, leading to frequent avalanche disasters, which seriously threaten people's lives and properties and the safety of transportation corridors in alpine mountainous areas. Taking the Daxiong River Basin downstream of the Yarlung Zangbo River as the research object, 70 avalanche points were identified and verified based on remote sensing interpretation and field investigation. Pearson correlation coefficient analysis was used for conducting covariance analysis, and 10 evaluation factors including elevation, slope, slope direction, ground curvature, surface cutting depth, surface cover type, vegetation coverage, TWI, average annual minimum temperature, and NDSI were comprehensively selected to construct an avalanche susceptibility evaluation system. The information value model was used for avalanche susceptibility zoning on the ArcGIS platform, dividing the study area into three categories: low susceptibility, medium susceptibility, and high susceptibility zones, and accuracy verification was conducted using ROC curve. The results show that the AUC value of the avalanche susceptibility evaluation model is 0.835, indicating good predictive accuracy. The areas of low, medium, and high susceptibility zones are 60.61 km², 74.33 km², and 96.91 km², respectively, accounting for 26.14%, 32.06%, and 41.80% of the total area of the study area. High susceptible zones are mainly located in mid-to-high and high-altitude areas, with Mount Duoxiongla, and Lage being typical. Finally, based on the combination of active and passive defense measures, integrated monitoring and early warning techniques, and corresponding architectural structures are proposed, providing technical support and scientific reference for disaster prevention and mitigation in the Duoxiong River Basin.

The loess region is characterized by complex geomorphological patterns. This region is prone to frequent earthquakes with serious seismic landslide disasters. Loess seismic landslides are affected by a variety of factors, including the topography and geomorphology of loess slopes, stratigraphic lithology, dynamic responses, strength and dynamic characteristics of loess, and hydrogeological conditions. Current research on loess seismic landslides primarily involves laboratory experiments, physical and numerical simulations, field investigations, and remote sensing and monitoring techniques. The research focuses on the mechanisms, development characteristics, distribution, dynamic responses, and stability of loess seismic landslides. This paper reviews the current state of both domestic and international research on loess seismic landslides, introduces the pseudo-dynamic method that considers seismic wave propagation characteristics, and outlines future research prospects based on this method. By analyzing the mechanics mechanisms of loess seismic landslide, investigating the seismic liquefaction phenomena of loess landslides, and discussing the instability characteristics of these landslides, this study proposes a calculation method to accurately evaluate the stability of loess seismic landslides. This research can provide a theoretical basis for earthquake disaster prevention and mitigation in loess areas, and it represents a key focus for future studies.

The upper reaches of the Yellow River, located on the northeastern edge of the tectonically active Qinghai Tibet Plateau, are characterized by complex geological conditions that have led to a high incidence of geological disasters such as landslides and collapses. This study focuses on the secondary landslide at the front part of the Xijitan landslide on the north side in Jiangla Village, Garang Township, Guide County, Qinghai Province. Using methods including unmanned aerial vehicle surveying, InSAR surface displacement monitoring, and on-site investigation methods, a detailed analysis of the geological environmental conditions, development characteristics, and reactivation mechanism of the landslide are conducted. On-site investigation results indicate that the rock mass structure in the landslide area is fragmented, with concentrated rainfall. The main exposed strata are Neogene mudstone and Holocene slope deposits. Multiple large cracks and tension bands are developed at the rear edge of the landslide, and the surface deformation is obvious, in the stage of creep deformation. Analysis of the mechanism of landslide reactivation reveals that densely developed structures play a controlling role in the reactivation of secondary landslides, and the softening of mudstone caused by concentrated rainfall is a key factor in inducing landslide reactivation. The interaction between two factors continuously reduce the integrity and strength of the rock mass, leading to the deformation and reactivation of secondary landslides. The research results aim to provide a theoretical basis for disaster prevention and reduction work in the upper reaches of the Yellow River.

Long-runout rockslides at high altitude have caused lots of severe casualties and huge economic losses in the world, becoming a focus issue in researches on mitigation for large-scale geological disasters. This paper systematically reviews the research process of high-altitude and long-runout rockslides and believes that conventional research on “high velocity and long runout” is difficult to adapt to the requirements of complex geohazards prevention and mitigation in high and extra-high mountains. The methodology on high-altitude and long-runout rockslides has been proposed that includes in the initiation at the high-position, the dynamics of chain-style disasters with a long-runout traveling and the risk assessment and mitigation. Then, the disaster-prone geostructure characteristics and early identification techniques of the high-altitude initiation zone, the long-runout transferring mechanism and boundary layer effect of high-velocity debris avalanche, and risk assessment and mitigation issues have been explored. Through the study in the high mountain and extra-high mountains of the Qinghai-Tibet Plateau indicates that the potential flow transferring mechanism of debris avalanche in high-altitude rockslides, the boundary layer effect of turbulent fluid and the plowing bodies. It is proposed that energy dissipation and risk mitigation methods can be used by modifying the boundary layer bottom slope of high potential debris avalanche, to increase the generation of turbulent kinetic energy in the boundary layer, and the dead zone range in front of barrier piles. Three research directions have been discussed, including the initiating mechanism of disaster-prone geostructure, the dynamic process of high-altitude and long-runout disaster chains, and the theory and technology of risk prevention and mitigation.

Studying the strength characteristics of the accumulation layer in ancient landslides is an important aspect of landslide stability analysis and control. In order to reasonably determine the strength parameters in the stability analysis of ancient landslide accumulation layers, this study focuses on the Jiangdingya ancient landslide treatment project in Zhouqu County, Gansu Province. The study targets the the sliding body soil (gravel soil) and sliding zone soil (gravelly clay) and conducts large-scale field shear tests at two typical sections using the horizontal pushing method. The test results are compared and analyzed with the suggested mechanical parameters based on engineering analogy method. Based on the above research, the following conclusions are drawn: (1) When the shear strain of the sliding zone soil is small, the shear stress-shear displacement relationship curve has crossed, and the shear stress of the soil sample increases with the increase of normal stress and shear strain; (2) The larger the normal stress of the sliding body soil, the more significant the strain softening behavior; (3) The residual strength of the landslide soil has a certain attenuation compared to the peak strength, mainly manifested as a weakening of cohesion, while the internal friction angle remains almost unchanged. This is because the cementitious material in the sliding zone soil is destroyed after shear deformation, leading to a weakening of cementation. The natural residual strength parameters of the sliding body soil on the discharge slope are: the internal friction angle ranges from 18.2° to 24.6°, and the cohesion ranges from 10.2 kPa to 12.4 kPa. The residual strength parameters of the sliding body soil are closer to the values obtained by the engineering analogy method; (4) The recommended values of rock mass mechanical parameters for the sliding zone soil and sliding body soil of the accumulation layer landslide at Jiangdingya are as follows : For the sliding zone soil, the cohesion ranges from 7.0 kPa to 14.8 kPa, and the internal friction angle ranges from 15.0° to 17.5°; for the sliding body soil, the cohesion ranges from 15.8 kPa to 30.9 kPa, and the internal friction angles from 23.9° to 24.4°.

The geological environment in southeast Xizang is complicated and disasters occur frequently. Landslides pose a great threat to engineering construction and human and financial safety in the region. In order to select models with higher precision for regional landslide prediction in southeast Xizang, this paper used modified landslide point data through field investigation, combined with topographic and geomorphic factors, geological factors, land cover factors and induced factors, and screened the factors through principal component analysis. Frequency ratio models, BP neural network models, and a combination of the two models were used for regional landslides prediction in southeast Xizang. Finally, ROC curves were used to evaluate the model accuracy. The results showed that the frequency ratio model after factor selection had the highest prediction accuracy for southeast Xizang (AUC=0.889). Models with factors removed through principal component analysis had higher accuracy than those without removal, and landslides in southeast Xizang were mainly distributed along river systems, including the Yarlung Zangbo River, Daqu River, Zangqu River, Nujiang River, Lancang River, Weiqu River, Janqu River and Zhaqu River. The models were used to predict the disaster in the study area, revealing that landslide points were located in high susceptibility and susceptibility zones. The models developed in this study can provide technical reference for engineering construction in southeast Xizang.

Extreme rainfall is often accompanied by mass geological disasters, which seriously endangers the safety of people 's lives and property in prone areas and affects the healthy development of the economy and society. Summarizing and analyzing the time-space distribution characteristics of geological disasters due to extreme rainfall and the effectiveness of early warning is of great significance for improving the comprehensive defense ability against geological disasters. Taking the sudden geological disasters caused by “23•7” heavy rainfall in 2023 as the research object, based on the refined precipitation data from the Beijing sudden geological disaster monitoring and early warning system, the time-space distribution characteristics of “23•7” heavy rainfall and the development and distribution characteristics of geological disasters were analyzed, and the early warning effect of geological disasters was discussed. The results show that the“23•7”heavy rainfall has the characteristics such as a large total amount, strong rainfall, long duration and wide range, and the disasters due to extreme rainfall have the characteristics of group occurrence. The multi-dimensional early warning of geological disaster classification has achieved remarkable results and has achieved the goal of zero casualties due to geological disasters under extreme weather conditions. The research results can provide a reference for actively preventing and scientifically responding to extreme rainfall geological disasters.

Heixiluo gully, located in Suxiong Town, Ganluo County of Sichuan Province, experienced a catastrophic debris flow disaster at 8:00 a.m. on August 31, 2020, causing significant losses to local residents, as well as to the Chengdu–Kunming railway bridges and infrastructure. To Study the activity and dynamic characteristics of the debris flow in Heixiluo gully, a comprehensive analysis was conducted using field investigations, on-site observations, and high-precision DEM data from the study area. The formation conditions and activity characteristics of the “8•31” debris flow were studied, and the Massflow software was utilized to simulate and verify the debris flow, inversely simulating the dynamic evolution process of the debris flow in Heixiluo Gully, and quantitatively evaluating the dynamic characteristics of the“8•31”debris flow. The study indicates that the “8•31” debris flow mainly underwent a“snowballing” cycle of processes, including “rainfall runoff convergence-, incision and erosion on the sides, sedimentation and selection on the first-level platform, erosion on the rear side of the platform, further sedimentation and selection on the second-level platform, further incision and erosion on the rear side of the platform, bank slope collapse, blockage and collapse, river blockage, formation of barrier lakes, and dam breach discharge.” Based on the Massflow analysis of the dynamic process of the debris flow, the simulated peak discharge, flow velocity, flow depth, erosion, and sedimentation depths in each gully segment match the measured data, confirming the reliability of this method. Through this method, the dynamic characteristics of debris flow can be more intuitively analyzed, providing a theoretical basis for subsequent disaster prevention and mitigation works.