TerraMosaic Daily Digest: Jan 12, 2026
Daily Summary
Today's landslide research highlights a diverse range of approaches, from advanced AI applications in hazard prediction and remote sensing to fundamental studies of soil mechanics and slope stability. A significant trend involves leveraging AI and machine learning for improved hazard assessment, including bushfire intensity prediction, landslide susceptibility mapping, and tunnel integrity evaluation. Remote sensing continues to play a crucial role, with new datasets and techniques for change detection, crater detection, and monitoring volcanic activity. Several papers address the impact of climate change on slope stability, flood properties, and alluvial fan dynamics, emphasizing the need for climate-resilient infrastructure and adaptation strategies.
Studies also focus on understanding the complex interactions between geological factors, such as faults and rock discontinuities, and environmental conditions, including rainfall, temperature, and freeze-thaw cycles, to better predict and mitigate landslide hazards. The research demonstrates an increasing emphasis on multi-scale analysis, from micro-level soil mechanics to regional-scale hazard mapping, with several papers exploring innovative methods for integrating multiple data sources and modeling approaches. Finally, several papers address the challenges of applying AI in geosciences, highlighting the need for human oversight and the importance of including diverse perspectives in research.
Key Trends
- AI-Driven Hazard Assessment: Machine learning is increasingly used for predicting bushfire intensity, landslide susceptibility, and tunnel integrity, demonstrating the potential for improved disaster preparedness and risk management.
- Remote Sensing Advancements: New datasets and techniques are emerging for change detection, crater detection, and monitoring volcanic activity, providing valuable data for hazard assessment and environmental monitoring.
- Climate Change Impacts: Studies are focusing on the effects of climate change on slope stability, flood properties, and alluvial fan dynamics, emphasizing the need for climate-resilient infrastructure and adaptation strategies.
Selected Papers
This digest features 76 selected papers from 2,675 papers analyzed across multiple journals. Each paper has been evaluated for its relevance to landslide research and includes links to the original publications.
1. Australian Bushfire Intelligence with AI-Driven Environmental Analytics
Core Problem: Accurate prediction of bushfire intensity for effective disaster preparedness and response in Australia.
Key Innovation: Integration of historical fire events, meteorological observations, and vegetation indices with machine learning models (Random Forest, XGBoost, LightGBM, MLP, and ensemble classifier) to predict bushfire intensity.
2. RainBalance: Alleviating Dual Imbalance in GNSS-based Precipitation Nowcasting via Continuous Probability Modeling
Core Problem: Highly imbalanced temporal distribution of precipitation data limits the performance of GNSS station-based precipitation nowcasting models.
Key Innovation: A continuous probability modeling framework (RainBalance) that uses a variational autoencoder (VAE) to map clustered input samples into a continuous latent space, alleviating the imbalance issue by modeling continuous probabilistic label distributions.
3. WHU-PCPR: A cross-platform heterogeneous point cloud dataset for place recognition in complex urban scenes
Core Problem: Existing point cloud datasets lack diversity in scenes, platforms, and sensors, limiting the effective development of place recognition research.
Key Innovation: A new cross-platform heterogeneous point cloud dataset (WHU-PCPR) collected from vehicle-mounted and portable laser scanning systems in complex urban and campus scenes over a 60-month period.
4. SourceNet: Interpretable Sim-to-Real Inference on Variable-Geometry Sensor Arrays for Earthquake Source Inversion
Core Problem: Inferring high-dimensional physical states from sparse, ad-hoc sensor arrays is complicated by irregular geometries and the profound Sim-to-Real gap in physical modeling.
Key Innovation: A Transformer-based framework (SourceNet) that treats the sensor array as a flexible set to model arbitrary geometries and Physics-Structured Domain Randomization (PSDR) to randomize the governing physical dynamics.
5. Towards Operational Streamflow Forecasting in the Limpopo River Basin using Long Short-Term Memory Networks
Core Problem: Lack of spatiotemporal observational data limits the application of deep learning models for hydrological discharge simulation in data-scarce regions of Africa, specifically the Limpopo River basin.
Key Innovation: Application of LSTM models to hydrological discharge simulation in the transboundary Limpopo River basin, assessing the impact of varying input data on performance and addressing the challenges of data scarcity and human influence on data-driven modeling.
6. From Landslide Conditioning Factors to Satellite Embeddings: Evaluating the Utilisation of Google AlphaEarth for Landslide Susceptibility Mapping using Deep Learning
Core Problem: Traditional landslide susceptibility mapping relies on landslide conditioning factors (LCFs), which can be heterogeneous, uncertain, and have limited availability, constraining mapping reliability.
Key Innovation: Evaluates Google AlphaEarth embeddings as a standardized, information-rich alternative to conventional LCFs for LSM, demonstrating improved performance across multiple regions and deep learning models. This is the Editor's Choice paper for its innovative approach to landslide susceptibility mapping, offering a potentially more robust and scalable method particularly in data-scarce regions.
7. OceanSAR-2: A Universal Feature Extractor for SAR Ocean Observation
Core Problem: Developing a universal feature extractor for SAR-based ocean observation that can perform well across various downstream tasks.
Key Innovation: Improved self-supervised learning and dynamic data curation strategies for a foundation model, OceanSAR-2, demonstrating strong transfer performance across geophysical pattern classification, wind/wave estimation, and iceberg detection.
8. Universal scaling between precursory duration and event size across mechanically driven geohazards
Core Problem: The duration of the precursory phase of catastrophic events (landslides, rockbursts, etc.) is poorly constrained across sites, scales, and hazard types, limiting early warning capabilities.
Key Innovation: Identifies a robust scaling relationship between precursory duration and failure volume for mechanically driven instabilities, suggesting common organizing mechanisms across different geohazards.
9. Vision-Language Model for Accurate Crater Detection
Core Problem: Reliable crater detection on lunar surfaces is challenging due to varying crater sizes/shapes and difficult imaging conditions, posing risks to safe lunar landings.
Key Innovation: Proposes a deep-learning CDA based on the OWLv2 model (Vision Transformer) fine-tuned with a combined loss function (CIoU and contrastive loss) to achieve reliable crater detection across challenging lunar imaging conditions.
10. Exchange Is All You Need for Remote Sensing Change Detection
Core Problem: Effective fusion and discrimination of bi-temporal features are crucial for remote sensing change detection, but prevailing methods use complex difference computation modules.
Key Innovation: Introduces SEED, a streamlined paradigm that replaces explicit differencing with parameter-free feature exchange, achieving state-of-the-art performance with a simpler, unified framework.
11. Leveraging Membership Inference Attacks for Privacy Measurement in Federated Learning for Remote Sensing Images
Core Problem: Evaluating privacy leakage in Federated Learning (FL) models applied to remote sensing image classification.
Key Innovation: Using Membership Inference Attacks (MIA) as a quantitative privacy measurement framework for FL in remote sensing, demonstrating that communication-efficient FL strategies reduce MIA success rates.
12. SC-MII: Infrastructure LiDAR-based 3D Object Detection on Edge Devices for Split Computing with Multiple Intermediate Outputs Integration
Core Problem: Deploying LiDAR-based 3D object detection models on edge devices with limited computational resources and addressing blind spots in single LiDAR setups.
Key Innovation: A split computing framework (SC-MII) that distributes processing between edge devices and a server, integrating multiple infrastructure LiDAR inputs to improve detection accuracy and reduce latency.
13. Terrain Diffusion: A Diffusion-Based Successor to Perlin Noise in Infinite, Real-Time Terrain Generation
Core Problem: Generating realistic and coherent infinite terrains in real-time is challenging due to limitations of traditional procedural noise functions.
Key Innovation: A generative framework called Terrain Diffusion that combines diffusion models with properties of procedural noise, enabling seamless infinite extent, seed-consistency, and constant-time random access for terrain generation.
14. Neighborhood Feature Pooling for Remote Sensing Image Classification
Core Problem: Enhancing texture-aware representation learning for remote sensing image classification.
Key Innovation: A novel Neighborhood Feature Pooling (NFP) layer that captures relationships between neighboring spatial features by aggregating local similarity patterns across feature dimensions.
15. Towards Stable Semi-Supervised Remote Sensing Segmentation via Co-Guidance and Co-Fusion
Core Problem: Alleviating the burden of exhaustive annotation in remote sensing image semantic segmentation while addressing the challenge of pseudo-label drift.
Key Innovation: A stable semi-supervised RS segmentation framework (Co2S) that synergistically fuses priors from vision-language models and self-supervised models, using a heterogeneous dual-student architecture and explicit-implicit semantic co-guidance.
16. CrackSegFlow: Controllable Flow Matching Synthesis for Generalizable Crack Segmentation with a 50K Image-Mask Benchmark
Core Problem: Limited pixel-level labels and domain shift hinder the deployment of computer vision-based inspection for infrastructure assets.
Key Innovation: CrackSegFlow, a controllable Flow Matching synthesis method that renders synthetic images of cracks from masks with pixel-level alignment, combined with topology-preserving mask injection and edge gating.
17. Dynamic and thermodynamic drivers of extreme precipitation under nonstationarity: implications for probable maximum precipitation across North America
Core Problem: Understanding the drivers of extreme precipitation events for improved risk assessment.
Key Innovation: Quantifies the contributions of precipitable water, convective available potential energy, and moisture divergence to extreme precipitation using a nonstationary framework.
18. Deciphering the role of ENSO and IOD in hydro-climate dynamics using SWAT and multi-dimension wavelet analysis
Core Problem: Assessing the impact of climate oscillations on hydro-climate variability in a fragile ecosystem.
Key Innovation: Establishes a teleconnection between ENSO/IOD and hydro-climate variables using wavelet analyses and projects hydrological extremes using the SWAT model.
19. Integrating policy measures into the assessment of household livelihood resilience to climate change in the Pumqu river Basin, Tibetan plateau
Core Problem: Evaluating household livelihood resilience to climate change, including the role of policy interventions.
Key Innovation: Develops a comprehensive framework integrating household endowment and policy interventions to assess livelihood resilience.
20. Impact of climate change on flood properties in a mountainous catchment of Nepal Himalayas
Core Problem: Assessing the impact of climate change on flood characteristics in the Nepal Himalayas.
Key Innovation: Analyzes changes in flood peak, volume, duration, and seasonality, highlighting the sensitivity of flood properties to precipitation and temperature changes.
21. Study on deformation mechanism and control technology of rock pillar in near-vertical coal seams mining
Core Problem: Mitigating dynamic disasters during mining of near-vertical coal seams.
Key Innovation: Investigates the mechanism of dynamic disasters induced by rock pillar deformation and optimizes pressure relief blasting schemes.
22. Model tests on slope failures caused by rising gas pressure
Core Problem: Understanding gas-induced slope failures, common in municipal solid waste landfills.
Key Innovation: Presents 1g model tests of slope failures caused by rising gas pressure, monitoring pore liquid and gas pressures independently.
23. Hydraulic conductivity and irrigation as controls on perched aquifer development and slope stability: A numerical case study from the Majes irrigation project, Peru
Core Problem: Investigating the influence of hydraulic conductivity and irrigation on perched aquifer development and slope stability.
Key Innovation: Uses numerical modeling to analyze the impact of irrigation practices on groundwater levels and slope stability in the Majes irrigation project.
24. Shape-Aware Topological Representation for Pipeline Hyperbola Detection in GPR Data
Core Problem: Enhancing the detection of underground utilities, especially pipelines, in GPR data.
Key Innovation: Integrates shape-aware topological features from GPR images using Topological Data Analysis (TDA) with the spatial detection capabilities of YOLOv5.
25. Digitalisation in geosciences for environmental protection
Core Problem: Applying AI to geosciences for environmental protection, including landslide mapping.
Key Innovation: Review of machine learning applications for soil, groundwater, landslides, and climate protection.
26. Long-term floor heave analysis of loess tunnel based on modified Green-Ampt infiltration model
Core Problem: Understanding and predicting floor heave in loess tunnels due to repeated rainfall infiltration.
Key Innovation: Improved Green-Ampt model for repeated infiltration, validated with 15-year tunnel survey data.
27. Deep learning-based inversion of stress curves and crack quantification in NPR anchored rock masses
Core Problem: Estimating internal mechanical parameters and fracture distribution in rock masses.
Key Innovation: Using deep learning to invert mechanical responses of NPR anchored fractured rock masses.
28. An advanced imaging scheme for fault structures utilizing gravity rank indicator
Core Problem: Interpreting gravitational responses from subsurface faults for resource exploration and hazard management.
Key Innovation: Novel inversion technique using local wavenumber and rank indicator for fault imaging.
29. Evaluation of volumetric water content in sand by centrifuge modelling of vertical rainwater infiltration
Core Problem: Evaluating slope stability under rainfall by assessing volumetric water content during infiltration.
Key Innovation: Centrifuge experiments with a rainfall simulator to quantify volumetric water content in sand.
30. A unified Cosserat-based anisotropy evolution FEM model and its application to the analysis of the San Francisco Bay subaqueous slope failure
Core Problem: Modeling slope stability in natural saturated clays considering strength anisotropy and strain softening.
Key Innovation: Unified anisotropic evolution equation with Cosserat continuum framework and a practical approach to determine internal length scale.
31. Experimental study on seismic response characteristics of coral sand and silt interbedded foundation
Core Problem: Understanding the seismic response of coral sand and silt interbedded foundations in reclamation projects.
Key Innovation: Shaking table tests comparing interbedded foundations to pure coral sand foundations.
32. Multi-stage Rotation-Stirring and Diameter-Expanding Integrated Strengthened Anchoring Technology for Large-Diameter Cable Bolts and Its Applications
Core Problem: Improving the anchoring efficiency of large-diameter cable bolts in soft surrounding rock.
Key Innovation: Multi-stage rotation-stirring and diameter-expanding integrated strengthened anchoring technology.
33. Experimental and Theoretical Study on the Dynamic Tensile Failure Mechanism of Freeze–Thaw Treated Rock Under Static Pre-Tension
Core Problem: Understanding the dynamic tensile mechanical characteristics of rocks under freeze-thaw and static-dynamic loads.
Key Innovation: Coupled static-dynamic loading tests on freeze-thaw treated rock specimens with multiple observation methods.
34. Interaction Between Strain-Softening Surrounding Rock and Support from an Energy Perspective
Core Problem: Understanding the energy transfer mechanism between surrounding rock and tunnel support in deep soft rock tunnels.
Key Innovation: Energy calculation method for surrounding rock based on the Mohr–Coulomb strain-softening model.
35. Experimental Study on the Weakening Effect and Mechanism of Strainburst by Boreholes with Directional Crack
Core Problem: Minimizing strainbursts in deep underground excavation using boreholes with directional cracks.
Key Innovation: True-triaxial uploading strainburst experiment to verify the weakening effect of boreholes with directional cracks.
36. Influence of Pilot Tunnel Excavation Sequence on Surface Subsidence in Metro Stations Constructed Using the Pile Beam Arch Method
Core Problem: Reducing surface subsidence during metro station construction using the pile-beam-arch method.
Key Innovation: Analyzing the optimal excavation sequence of pilot tunnel groups based on surface subsidence control.
37. Creep of water-bearing soft rock and its influence on long-term rock mass stability
Core Problem: Understanding the long-term stability of rock masses affected by creep in water-bearing soft rock.
Key Innovation: Investigating the creep behavior of water-bearing soft rock and its impact on rock mass stability.
38. Distribution-free estimation for three-dimensional diameter of rock discontinuities within complex high-steep slope based on Bertrand paradox
Core Problem: Estimating the size and distribution of rock discontinuities in complex, steep slopes.
Key Innovation: Using Bertrand paradox to estimate the 3D diameter of rock discontinuities.
39. Creep behavior and microstructural evolution of unsaturated red-bed mudstone under coupled matric suction-stress effects
Core Problem: Understanding the creep behavior of unsaturated red-bed mudstone.
Key Innovation: Analyzing creep behavior under coupled matric suction-stress effects.
40. Factors Influencing Return Migration after Disaster Relocation: Housing Dissatisfaction and Place Attachment in the 2021 Mount Semeru Eruption, Indonesia
Core Problem: Understanding factors influencing return migration after disaster relocation due to the 2021 Mount Semeru eruption.
Key Innovation: Analyzing housing dissatisfaction and place attachment in the context of debris flow hazards.
41. Preferential flow paths in active rock glaciers
Core Problem: Understanding water flow within rock glaciers is crucial for predicting their stability and response to climate change.
Key Innovation: Review of preferential flow paths in active rock glaciers, contributing to understanding their internal dynamics.
42. 2023 activity of Nyamulagira volcano monitored by SAR interferometric coherence
Core Problem: Monitoring volcanic activity is crucial for hazard assessment and risk management in volcanically active regions.
Key Innovation: Using SAR interferometric coherence to monitor volcanic activity.
43. Rill and interrill erosion on slopes in granite red soil quantified by using Sm and Eu as tracers: An indoor rainfall simulation study
Core Problem: Quantifying soil erosion processes on slopes is essential for understanding land degradation and developing effective soil conservation strategies.
Key Innovation: Using Sm and Eu as tracers to quantify rill and interrill erosion in rainfall simulation experiments.
44. Decoupling effects of driving factors on sediment load in the middle reaches of the Yellow River
Core Problem: Understanding the factors influencing sediment load in rivers is crucial for managing water resources and mitigating soil erosion.
Key Innovation: Analyzing the decoupling effects of various driving factors on sediment load in the Yellow River.
45. Development and application of a distributed hydrology and soil erosion model in a semi-arid catchment, China
Core Problem: Soil erosion in semi-arid catchments leads to land degradation and reduced agricultural productivity.
Key Innovation: Development and application of a distributed hydrological and soil erosion model for a semi-arid catchment.
46. Integrated land preparation and afforestation enhance soil erosion resistance in the hilly and gully region of the Loess Plateau, China
Core Problem: Soil erosion in the Loess Plateau.
Key Innovation: Integrated land preparation and afforestation techniques to enhance soil erosion resistance.
47. Performance evaluation of strain gauges in frozen soils and application to load-transfer analysis of steel pipe piles
Core Problem: Evaluating the performance of strain gauges in frozen soils.
Key Innovation: Application of strain gauges for load-transfer analysis of steel pipe piles in cold regions.
48. Hydromechanical analysis of shear behaviors of sandstone with ice-filled fractures
Core Problem: Shear behavior of sandstone with ice-filled fractures.
Key Innovation: Hydromechanical analysis of shear behavior in fractured sandstone under cold conditions.
49. Efficiency and environmental benefits of road snow-melting, anti-icing and icing smart detection technologies in cold regions: Review and discussion
Core Problem: Road safety in cold regions.
Key Innovation: Review of snow-melting, anti-icing, and icing smart detection technologies.
50. Coupled effects of subzero temperature and impact loading on the dynamic mechanical properties and fracture mechanism of dolomite
Core Problem: Fracture mechanism of dolomite under subzero temperatures and impact loading.
Key Innovation: Analysis of dynamic mechanical properties and fracture behavior of dolomite in cold regions.
51. Failure mechanism and stability control of fault induced by underground cavern excavation: insights from theoretical analysis and numerical modeling
Core Problem: Stability of faults induced by underground cavern excavation.
Key Innovation: Theoretical analysis and numerical modeling to understand failure mechanisms and control stability.
52. Advantages of geopolymers in rock joint grouting reinforcement: experimental study and peak shear strength model
Core Problem: Rock joint reinforcement is crucial for slope stability and tunnel support, but traditional methods have limitations.
Key Innovation: Using geopolymers for rock joint grouting to improve shear strength and stability.
53. A real-time prediction model for tunnel rock strength using geological drilling data and physics-informed neural networks
Core Problem: Accurate and timely prediction of rock strength is essential for safe and efficient tunnel construction.
Key Innovation: A physics-informed neural network model for real-time rock strength prediction using geological drilling data.
54. Evaluation of tunnel rock mass integrity using multi-modal data and Generative Large Language Model: TunnelRIP-GPT
Core Problem: Assessing rock mass integrity in tunnels is critical for safety and stability, but current methods can be limited.
Key Innovation: Using a Generative Large Language Model (TunnelRIP-GPT) with multi-modal data for tunnel rock mass integrity evaluation.
55. Water inrush disasters behavior in weakly cemented red-bed tunnels using 3DEC–Flow coupled model
Core Problem: Water inrush in tunnels, especially in weakly cemented red-beds, poses significant safety risks.
Key Innovation: Using a 3DEC-Flow coupled model to simulate and understand water inrush behavior in tunnels.
56. Deformation characteristics and progressive failure mechanism of layered carbonaceous slate tunnel based on a discrete–continuum coupling method
Core Problem: Understanding the deformation and failure mechanisms of tunnels in layered rock formations is crucial for stability.
Key Innovation: Applying a discrete-continuum coupling method to analyze the deformation and progressive failure of tunnels in layered carbonaceous slate.
57. Unsupervised deep-learning model for quantitative diagnosis of tunnel lining cavity condition using interfering GPR data
Core Problem: Diagnosing tunnel lining cavity conditions is essential for maintenance, but GPR data interpretation can be challenging.
Key Innovation: An unsupervised deep-learning model for quantitative diagnosis of tunnel lining cavities using GPR data.
58. Seismic response characteristics and instability mechanism of strongly weathered sandstone high-steep slopes containing two tunnels
Core Problem: Understanding the seismic response and instability mechanisms of high-steep slopes made of strongly weathered sandstone, particularly those containing tunnels.
Key Innovation: Analysis of seismic response and instability of weathered sandstone slopes with tunnels.
59. Grain crushing and response of sands and slopes along slip surfaces until large sliding displacement
Core Problem: Investigating grain crushing and its effect on the response of sands and slopes along slip surfaces during large sliding displacements.
Key Innovation: Study of grain crushing effects on slope stability and sand behavior during sliding.
60. Stochastic assessment of vertical soil heterogeneity on seismic performance and vulnerability in mildly liquefiable sloping ground with stone columns
Core Problem: Assessing the impact of vertical soil heterogeneity on the seismic performance and vulnerability of mildly liquefiable sloping ground improved with stone columns.
Key Innovation: Probabilistic assessment of soil heterogeneity effects on liquefaction in sloping ground.
61. Assessing liquefaction potential using Arias intensity: a logistic regression approach with an updated database
Core Problem: Improving the assessment of liquefaction potential by using Arias intensity and a logistic regression approach with an updated database.
Key Innovation: Liquefaction assessment using Arias intensity and logistic regression.
62. Dynamic response of soil-rock mixture embankments under coupled cyclic loading and seepage: A DEM-FDM approach
Core Problem: Understanding the dynamic response of soil-rock mixture embankments under coupled cyclic loading and seepage conditions.
Key Innovation: Analysis of soil-rock embankment response using DEM-FDM coupling.
63. An approach for assessing seismic response of degrading permafrost sites
Core Problem: Developing an approach for assessing the seismic response of sites with degrading permafrost.
Key Innovation: Seismic response assessment for degrading permafrost sites.
64. Insights into seismic interactions of three aligned structures on layered liquefiable ground
Core Problem: Investigating the seismic interactions of three aligned structures on layered ground susceptible to liquefaction.
Key Innovation: Analysis of seismic interactions of structures on liquefiable ground.
65. Probabilistic assessment of seismic settlement of embankment constructed on soft soil by combining simulation of stochastic ground motion and spatial soil variability
Core Problem: Assessing the seismic settlement of embankments on soft soil, considering stochastic ground motion and spatial soil variability.
Key Innovation: Probabilistic assessment of embankment settlement considering ground motion and soil variability.
66. Pixel-level landslide stability mapping based on three-dimensional time-series InSAR monitoring and dynamics factors extraction
Core Problem: Landslide stability mapping using InSAR data is challenging due to complex terrain and dynamic factors.
Key Innovation: Pixel-level landslide stability mapping using 3D time-series InSAR and dynamic factors extraction.
67. Blasting vibration analyses of open-pit mine slope
Core Problem: Blasting vibrations can induce slope instability in open-pit mines.
Key Innovation: Analysis of blasting vibration effects on open-pit mine slope stability.
68. Improved analysis method for frame beams with prestressed cables in slopes based on inclined foundation stiffness
Core Problem: Analyzing frame beams with prestressed cables in slopes requires accurate modeling of foundation stiffness.
Key Innovation: Improved analysis method for frame beams in slopes considering inclined foundation stiffness.
69. Prediction of long-term deformation of slope in Baihetan hydropower station based on updated rheological parameters with Bayesian estimation
Core Problem: Long-term slope deformation prediction is crucial for hydropower station safety.
Key Innovation: Slope deformation prediction using updated rheological parameters with Bayesian estimation.
70. Bearing capacity, shear band evolution, and deformation characteristics of slopes reinforced by root-inspired anchors using transparent soil model testing
Core Problem: Understanding the behavior of slopes reinforced with root-inspired anchors.
Key Innovation: Transparent soil model testing to analyze slope reinforcement with root-inspired anchors.
71. Slope rockbolting using key block theory: Force transfer and artificial intelligence-assisted multi-objective optimisation
Core Problem: Optimizing rock bolting design for slope stabilization.
Key Innovation: AI-assisted multi-objective optimization of rock bolting design using key block theory.
72. Climate‐driven alluvial fan aggradation and incision in an unglaciated Himalayan basin, Northwestern India
Core Problem: Understanding the impact of climate change on alluvial fan dynamics in the Himalayas.
Key Innovation: Analysis of alluvial fan aggradation and incision patterns in relation to climate variability in a specific Himalayan basin.
73. Spatial and morphometric analysis of a comprehensive dataset of loess sinkholes from a small basin in the Chinese Loess Plateau
Core Problem: Loess sinkholes pose a significant hazard in the Chinese Loess Plateau, but their spatial distribution and morphometric characteristics are not well understood.
Key Innovation: Comprehensive dataset of 1,194 loess sinkholes mapped using airborne and handheld laser scanners, providing high-resolution data for analysis and management.
74. Study on the deformation mechanism of a fault-dominated creep-sliding and tension-fracturing slope in a dam site area
Core Problem: Understanding the deformation mechanism of a slope influenced by faults, creep-sliding, and tension-fracturing in a dam site area.
Key Innovation: Analysis of slope deformation mechanism combining fault influence, creep-sliding, and tension-fracturing.
75. Beyond structured knowledge: performance boundaries of ChatGPT in geological-hazard question answering and the need for human-in-the-loop oversight
Core Problem: Evaluating the performance of ChatGPT-4o in answering questions related to geological hazards and identifying its limitations.
Key Innovation: Rubric-based evaluation of a large language model (ChatGPT-4o) for geohazard question answering, highlighting the need for human oversight.
76. African contributions are missing from cryosphere research in Africa and worldwide
Core Problem: Addressing the lack of African contributions to cryosphere research, particularly concerning snow and ice avalanches in Africa.
Key Innovation: Highlights the need for greater inclusion of African researchers and perspectives in cryosphere studies.