TerraMosaic Daily Digest: Feb 6, 2026
Daily Summary
This digest synthesizes 74 selected papers and focuses on landslide process mechanics and slope evolution, flood generation, routing, and hydroclimatic forcing, coastal and submarine hydro-geomechanics. Top-ranked studies examine mass-movement initiation and runout dynamics, earthquake-triggered slope response and liquefaction, and flood generation and hydroclimatic forcing.
Across the full set, evidence converges on mechanism-constrained analysis with operational relevance, especially for seismic source-to-ground response pathways and infrastructure-focused hazard performance. The strongest contributions pair interpretable process evidence with monitoring or forecasting workflows that support warning design and risk prioritization.
Key Trends
- Landslide studies increasingly resolve process chains: Contributions connect triggering conditions, slope deformation, and mobility outcomes, improving the basis for warning thresholds and scenario testing.
- Flood analyses are becoming event-specific and process-based: Papers emphasize precipitation structure, antecedent wetness, and catchment controls rather than static hazard descriptors.
- Coastal and submarine hazards are treated as coupled systems: Wave, mass-transport, and shoreline processes are analyzed together with engineering implications.
- Seismic hazard research links source behavior to ground response: Recurring topics connect rupture or loading conditions with geotechnical performance and consequence assessment.
- Infrastructure-facing outputs are increasingly decision-ready: Asset performance is evaluated with uncertainty-aware frameworks to support mitigation and maintenance prioritization.
Selected Papers
This digest features 74 selected papers from 567 deduplicated papers analyzed (out of 2145 raw papers scanned). Each paper has been evaluated for its relevance to landslide and broader geohazard research and includes links to the original publications.
1. The Shaweitaizi river-blocking landslide and its impacts on landscape evolution in the downstream of the Jinsha River
Core Problem: Understanding the causes, geomorphological and sedimentological impacts, and landscape evolution resulting from the Shaweitaizi landslide-induced river damming along the Jinsha River.
Key Innovation: Investigates the Shaweitaizi landslide, identifying contributing factors (stratigraphy, seismicity, dip-slope), revealing a secondary landslide, characterizing upstream lacustrine sediments (thickness, elevation, chronology via 14C and OSL), calculating the extensive inundation of the landslide-dammed lake (338 km backwater), and inferring catastrophic flood event as the cause of dam failure, providing insights into geohazard processes in tectonically active fluvial environments.
2. Cycled Fluid Injection Limits Maximum Earthquake Size by Controlling the Cadence of Seismic Moment Release
Core Problem: Debate exists on whether and how controlling fluid injection can limit the size of injection-induced seismicity.
Key Innovation: Laboratory experiments show that cycled fluid injection, compared to continuous injection, triggers more frequent but smaller seismic events, reducing maximum moment magnitude and deformation energy by actively reducing pore pressure and temporally partitioning slip events, suggesting a method for effective hazard mitigation.
3. Comparing Synoptic Pattern Evolution for Flash‐Flood‐Producing and Non‐Flash‐Flood‐Producing Mesoscale Convective Systems in the United States
Core Problem: Understanding how the short-term evolution of synoptic weather patterns influences Mesoscale Convective Systems (MCSs) to produce flash floods, which are responsible for over half of central U.S. flash floods.
Key Innovation: Used machine learning to analyze MCS data and atmospheric reanalyses, finding that while synoptic patterns reflect seasonal/regional differences, a broader precipitating area and strong water vapor transport driven by a nearby synoptic-scale forcing are the most dominant factors governing MCS flash flood potential.
4. Experimental study of wave-induced seabed response around a jacket structure with different arrangements
Core Problem: The safety of offshore structures (e.g., wind turbine foundations) is threatened by wave-induced seabed response and instability, particularly transient liquefaction, and the influence of structure arrangements on this phenomenon is not fully understood.
Key Innovation: Experimentally investigated wave-induced seabed response around jacket foundations, revealing specific pore pressure distribution patterns, the influence of wave conditions and structure arrangements (e.g., increased liquefaction potential with asymmetrical arrangements), providing insights into seabed instability mechanisms.
5. Storylines for the 1997 New Year’s Flood: The role of watershed antecedent conditions and future warming in shaping discharge in the Truckee River watershed
Core Problem: Understanding the complex interactions of flood drivers (extreme precipitation, wet antecedent conditions, warm temperatures, rapid snowmelt) and how future warming will shape extreme flood events, particularly for historical events like the 1997 New Year’s Flood, remains a challenge.
Key Innovation: Leveraging RRM-E3SM simulated forcings and a process-based hydrological model to recreate the 1997 New Year’s flood under various warming levels, demonstrating that wet antecedent watershed conditions are critical for reproducing such events and showing increased peakflows under future warming due to enhanced snowmelt, using a storyline approach.
6. Research on catastrophic process of large-scale slope using 3D-DDA with machine learning-based parameter optimization
Core Problem: Simulating the entire catastrophic process of large-scale slope instability using 3D Discontinuous Deformation Analysis (DDA) is challenging due to low computational efficiency and complex, subjective parameter determination.
Key Innovation: Enhances 3D DDA for large-scale landslide deduction by integrating OpenMP parallel computing and a machine learning-based surrogate model (LightGBM with Bayesian optimization) for parameter optimization. This approach accurately reproduces the catastrophic process and kinematic characteristics of a large-scale landslide, significantly improving the capability and efficiency of 3D DDA for such simulations.
7. Numerical investigation of pore pressure evolution mechanisms in saturated sand during liquefaction using coupled discrete element method
Core Problem: A comprehensive understanding of the fundamental physical mechanisms of pore pressure evolution in saturated sand during liquefaction, particularly the coupled effects of pore structure evolution and the relative contributions of porosity-change-induced (PI) and diffusion-induced (DI) mechanisms, is lacking.
Key Innovation: Develops a 3D fluid-particle coupling numerical model based on DEM with high-precision dynamic calculation of particle porosity, comprehensively coupling PI and DI pressurization/depressurization mechanisms. This model successfully simulates the complete liquefaction process, accurately reproduces key phenomena, and quantitatively separates the relative contributions of PI and DI mechanisms, revealing their synergistic control over different stages of liquefaction.
8. Brittle failure evaluation of rocks under freeze–thaw cycles based on an energy-based method
Core Problem: Existing methods for evaluating rock brittleness under freeze-thaw (F-T) cycles in cold regions are limited, despite its importance for slope stability and infrastructure safety.
Key Innovation: Proposed a novel energy-based brittleness index (BI) that comprehensively considers energy evolution throughout all stages of rock failure, validated its effectiveness across different rock types, and demonstrated its superior sensitivity to F-T cycles and confining pressure, providing valuable support for engineering assessment in cold regions.
9. Kinematic response of end-bearing piles in nonhomogeneous unsaturated soils subjected to seismic P-waves
Core Problem: Accurately predicting the kinematic response of end-bearing piles in complex, nonhomogeneous unsaturated soils when subjected to vertically propagating seismic P-waves is challenging due to the coupled soil-pile interaction and multi-phase soil behavior.
Key Innovation: Development of a novel analytical model based on three-phase poroelastic theory and Laplace transforms to investigate the kinematic response of end-bearing piles in nonhomogeneous unsaturated soils under seismic P-waves, providing frequency-domain solutions and insights into the influence of various soil and pile parameters on seismic response.
10. Local Land‐Atmosphere Interactions Precondition Moist and Dry Heatwaves Under Large‐Scale Subsidence Over the Indo‐Gangetic Plains
Core Problem: Understanding the formation mechanisms of pre-monsoon heatwaves over the Indo-Gangetic Plains, particularly the role of local land-atmosphere interactions versus large-scale anticyclones, to improve early warning systems.
Key Innovation: Demonstrated that local land-atmosphere interactions (diabatic and adiabatic processes, antecedent soil moisture, pre-monsoon showers, nocturnal low-level clouds) are critical preconditions for both moist and dry heatwaves, distinguishing heatwave regions from non-heatwave regions, and highlighting their importance for reliable early warning systems.
11. Salt Precipitation‐Driven Rock Failure Mode Transition During Geological CO2 Sequestration
Core Problem: Poor understanding of mechanical behavior and failure mechanisms associated with salt precipitation in drying zones during geological CO2 sequestration, particularly the transition of failure modes and its impact on reservoir stability and microseismicity.
Key Innovation: Experimental demonstration of significant deterioration in rock mechanical performance and, for the first time, observation of a distinct transition of failure mode from shear-to-tensile-dominated under uniaxial compression due to salt precipitation, providing critical insights into reservoir stability and injection-induced microseismicity.
12. Reduced Upwind Moisture Transport Contributes to Drought in the Agro‐Pastoral Ecotone of Northern China
Core Problem: Unclear mechanisms of how changes in moisture supply drive droughts in the Agro-Pastoral Ecotone of Northern China (APENC), making ecological restoration and food security vulnerable under warming.
Key Innovation: Quantification of APENC's precipitation moisture sources and trends using a moisture-tracking model, revealing that reductions in moisture inflow from key terrestrial source regions (East Asia, South Asia–Indian Ocean) combined with reduced local humidity primarily trigger and sustain drought severity.
13. Rheological properties and scour resistance of solidified slurry for offshore wind monopile protection
Core Problem: Protecting offshore wind monopile foundations from scour requires effective solidified slurry, but the relationships between slurry composition, rheological properties, and scour resistance are not fully understood, hindering optimal formulation.
Key Innovation: Systematically investigated the rheological properties and scour resistance of sludge-based solidified slurry, identifying optimal HPMC content and the diminishing returns of sodium silicate and low water-to-solid ratios, and proposed a dimensionless Scour Resistance Coefficient (SRC) to guide practical formulation for scour protection.
14. A review of hospital seismic resilience enhancement strategies and technologies: State of the art, challenges, and future
Core Problem: Lack of clearly defined, scientifically grounded technical pathways for enhancing hospital seismic resilience, despite its critical importance for post-earthquake emergency response.
Key Innovation: Proposes an enhancement-oriented framework integrating structural, nonstructural, and organizational strategies for hospital seismic resilience, considering "resource-service-management" demands and moving from passive assessment to proactive enhancement.
15. Decoding surface and root-zone soil moisture dynamics for agricultural drought assessment using multi-source climate records (1990–2019)
Core Problem: Accurate monitoring and prediction of agricultural droughts are essential, but the dominant factors affecting soil moisture variations at different depths are not fully explored, and satellite-based soil moisture climatological records are underutilized for drought assessment.
Key Innovation: Investigation of 30-year surface and root-zone soil moisture dynamics using ESA-CCI records to characterize agricultural drought events, revealing distinct responses of SSM and RZSM to precipitation and infiltration, and developing a novel knowledge-guided machine learning model for improved agricultural drought prediction.
16. Anatomy of geostructural response and failure uncertainty with IAA
Core Problem: Quantifying the uncertainty in the dynamic response and estimating the failure capacity of geostructures (e.g., embankment dams) under dynamic loads, especially considering epistemic uncertainties and computational demands.
Key Innovation: A framework based on Intensifying Artificial Acceleration (IAA) to expedite uncertainty quantification in nonlinear transient simulations, generating failure fragility curves influenced exclusively by epistemic uncertainties, and facilitating integration with performance-based earthquake engineering.
17. Crack density estimation in rock structures using machine learning techniques
Core Problem: Direct field measurement of crack density, a critical indicator for rock structure stability, is often impractical, necessitating alternative, efficient predictive approaches.
Key Innovation: Development and application of machine learning techniques, using physical properties consistent with Biot theory, to accurately estimate rock crack density, demonstrating improved performance with oversampling algorithms and identifying compressional and shear wave velocities as key predictive parameters for reliable stability evaluation.
18. A new dilation angle model and a cohesion-weakening and friction-weakening model for simulating post-peak deformation of brittle hard rocks
Core Problem: Accurately simulating the nonlinear post-peak deformation of brittle hard rocks, which involves complex strain-softening, dilatancy, and plastic flow, is challenging with existing rock constitutive models.
Key Innovation: Proposal of a novel segmented confinement and plastic-shear-strain-dependent dilation angle model and a cohesion-weakening and friction-weakening (CWFW) model, which, when implemented in FLAC, accurately capture the post-peak strain-softening and dilatancy deformation behaviors of brittle hard rocks.
19. An IIVY-AdaBoost prediction model for rockburst intensity: Integrating multi-strategy optimization with SHAP-driven insight
Core Problem: Accurate prediction of rockburst intensity is challenging due to imbalanced datasets, leading to biased or overfitted machine learning models.
Key Innovation: Development of an IIVY-AdaBoost machine learning framework, integrating SMOTE for class imbalance and an improved IVY algorithm for hyperparameter optimization, achieving high accuracy and robustness in rockburst intensity prediction, with SHAP analysis identifying key predictors.
20. Enhanced Adiabatic Heating Drives Faster Warming of Early Summer Hot Extremes in North China
Core Problem: Understanding the drivers behind the disproportionately faster warming of early summer hot extremes in North China since 1990, which leaves populations underprepared.
Key Innovation: Identified enhanced adiabatic heating, contributed by more frequent arrival and intensified descent of upper-level air particles from west and south, as the main driver for the faster warming of early summer hot extremes, linked to the prevalence of a ridge pattern over North China.
21. Investigating Deep Learning Knowledge Transfer in Streamflow Prediction From Global to Local Catchment
Core Problem: Accurate streamflow prediction is critical for flood forecasting and water resource management, particularly in data-scarce regions where traditional models struggle.
Key Innovation: Evaluated transfer learning (TL-LSTM) approaches for streamflow prediction, demonstrating that pre-training LSTMs on data-rich regions and fine-tuning with limited local data significantly improves prediction accuracy, even for 'ungauged' basins, advancing cross-basin generalization for hydrological prediction.
22. Euler-pole clustering of GNSS velocities using unsupervised machine learning in the Southeastern Tibetan Plateau: Crustal block identification and the dominance of sinistral-slip faults
Core Problem: Contradictory results in constraining fault slip rates and crustal block geometries in the Southeastern Tibetan Plateau (SETP) due to complex geodynamics, deformation patterns, and subjective choices of crustal block boundaries.
Key Innovation: Employment of an unsupervised machine learning Euler pole clustering algorithm to automatically resolve rigid crustal blocks using GNSS velocity vectors. The optimal clustering results redefine the first-order kinematics of the SETP, identifying 4 elongated crustal blocks delineated by arcuate sinistral-slip faults, thereby elucidating the dominance of these faults.
23. Study on non-convexity for Zhang-Zhu strength criterion based on microfracture mechanics
Core Problem: The Zhang-Zhu (ZZ) strength criterion, used for deep rock failure, exhibits non-smoothness and non-convexity in its failure envelope, with the underlying physical mechanism for this non-convexity requiring further exploration.
Key Innovation: Derivation of a micro-Zhang-Zhu (micro-ZZ) strength criterion from microfracture mechanics, establishing a correlation between microfracture mechanisms (e.g., microcrack density) and the macroscopic non-convexity of the ZZ criterion, and identifying dominant parameters influencing non-convexity.
24. Sensitivity and scale dependence of discretization and roughness in the hydrodynamic modeling of surface runoff caused by torrential rainfall
Core Problem: In hydrodynamic surface runoff simulations for flash flood risk assessment, the lack of observations for model calibration and understanding the sensitivity to specific model parameters (discretization, roughness) are crucial challenges for reliable planning.
Key Innovation: Analyzes how surface discretization and roughness affect surface runoff generation and depression storage in a hydrodynamic 2D-model in an alpine region, comparing five discretization methodologies and seven roughness parameterizations, providing guidance on model setups for ungauged basins and highlighting the impact of resolution on runoff and artificial depressions.
25. Cross-sectional average velocity predictions for double-layered vegetated open channels incorporating vegetation sheltering and blockage effects
Core Problem: Accurately predicting cross-sectional average velocity in open channels with double-layered vegetation is crucial for evaluating flood discharge capacity, but existing models may lack accuracy or computational efficiency.
Key Innovation: Development of a genetic programming (GP) based predictive model for cross-sectional average velocity, which innovatively incorporates vegetation sheltering and blockage effects, balancing accuracy and computational efficiency for flood discharge capacity assessment.
26. Groundwater responses to decadal rainfall variability in semi-arid South Africa
Core Problem: Groundwater responses to climate trends and variability in semi-arid South Africa are underexplored, despite groundwater being a vital resource and the region experiencing significant rainfall variability, hindering future water availability assessments and geohazard understanding.
Key Innovation: Analysis of long-term rainfall, temperature, and groundwater level data (1940-2022) in a semi-arid South African catchment, revealing intensified rainfall patterns and a strong decadal-scale dependence of groundwater levels on antecedent rainfall, advancing the conceptual understanding of groundwater responses to large-scale climatic patterns.
27. Natural biogenic coral gravel soils in coastal areas in southern China: Sampling, mechanical characterization, and microstructure observation
Core Problem: The unique geotechnical behavior of natural biogenic coral gravel soils (CGS) in coastal areas is largely enigmatic due to their distinct properties (biological origin, irregular shapes, porosity) and the extreme difficulty in collecting high-quality undisturbed samples.
Key Innovation: Successful upgrade of the Mazier sampling technique using a chemical drilling additive to collect high-quality undisturbed CGS specimens, enabling comprehensive mechanical characterization and microstructural observation, and establishing foundational datasets to unveil the mechanisms of soil shear strength in these unique biogenic soils.
28. The southwest Kalahari dune field does not emit dust post‐fire despite a lack of vegetation and above‐threshold winds
Core Problem: Understanding why the southwest Kalahari dune field, despite post-fire de-vegetation and fine-grain components, does not significantly emit dust, and identifying the erodibility controls.
Key Innovation: Findings suggest low dust emission post-fire due to high burned debris cover and surviving biological soil crusts, which protect the surface. It identifies low wind speeds, high initial surface cover, and biocrusts as limiting factors, but notes potential for future dust emission under drought and high grazing.
29. Field testing and fracturing mechanism of coal seam fracturing driven by high-pressure gas combustion
Core Problem: Challenges exist in pressurizing high-pressure gas for coal seam fracturing and understanding the fracturing mechanism to effectively enhance gas desorption and methane concentration.
Key Innovation: Developed a combustion-driven high-pressure air fracturing apparatus, conducted field testing with microseismic monitoring, and established a high-pressure air-water coupled coal-rock fracturing model, demonstrating effective pre-cracking radii and enhanced gas desorption.
30. Scour mechanism and efficiency contrast between under-expanded and expanded air jets in underwater
Core Problem: Lack of understanding regarding how air jet expansion state governs the mechanism and performance of seabed scouring, particularly for efficient operational mode selection in marine engineering applications like trenching.
Key Innovation: Experimentally identified two distinct scour mechanisms (viscous shear erosion for expanded jets, bearing-capacity failure for under-expanded jets) and quantified the superior efficiency of fully expanded jets in terms of particle entrainment, area, and diffusion rate, providing a quantitative basis for mode selection.
31. Efficient recognition of cone karst landforms through deep learning: insights from multi-source data fusion in southwest China
Core Problem: Accurate and efficient mapping of cone karst hills, a typical landform in tropical-subtropical karst landscapes, which is challenging with conventional methods due to complex topography.
Key Innovation: Develops an advanced deep learning approach for cone karst hill identification, integrating multi-source data (DEM, slope, local relief, spectral) with U-Net and DeepLab V3+ architectures, demonstrating U-Net's superior performance and establishing a scalable, transferable workflow for high-precision, large-scale mapping.
32. Influences of vegetation distribution on soil organic carbon accumulation and stability in a coastal wetland, Southeast China
Core Problem: Further investigation is needed into the underlying mechanisms by which vegetation distribution in coastal wetlands influences soil organic carbon (SOC) accumulation and stability.
Key Innovation: Systematically analyzed SOC accumulation and stability along a sea-land vegetation gradient, identifying Spartina alterniflora as having the highest SOC, recalcitrant organic carbon (ROC), and mineral-associated organic carbon (MAOC). Identified soil total nitrogen, belowground biomass, clay/silt content, and carbon-fixing bacteria as key drivers, providing a basis for coastal blue carbon management.
33. Numerical investigation of pipeline upheaval buckling in rockfills: significance of particle scale effect
Core Problem: Current semi-empirical design guidelines for uplift resistance of subsea pipelines in coarse-grained materials (rockfills) often neglect the significant impact of the particle scale effect (D/d50 ratio), leading to uncertainties in predictive models for pipeline upheaval buckling.
Key Innovation: Performs 3D DEM numerical analyses using elongated particle clumps to accurately capture rockfill morphologies, investigating pipeline upheaval buckling with various embedment and D/d50 ratios. The study demonstrates the significant particle scale effect, identifies a transition value for D/d50 (7.0-10.0), and proposes a two-parameter model and design trendlines for uplift resistance, improving predictive capabilities for infrastructure design.
34. Small-scale test on the response of adjacent piles caused by shield tunnel excavation in sand
Core Problem: Shield tunnel excavation causes additional settlement and stress on adjacent pile-supported structures, and there is a need to evaluate the influence of tunneling on these structures and surrounding soil to guide disaster prevention and control.
Key Innovation: Developed a small-scale testing machine to precisely simulate shield tunneling in sand, revealing the influence of pile length, diameter, pile-tunnel relative distance, and pile head load on surrounding soil displacement and adjacent pile axial force, providing scientific guidance for disaster prevention during tunnel excavation.
35. Modeling intrusion of bentonite buffer into fractures under the BExM framework: incorporation of wall friction effects
Core Problem: Existing models for bentonite intrusion into fractures in repositories are based on empirical relationships and ignore wall friction, leading to significant discrepancies in assessing intrusion behavior and potential radionuclide escape.
Key Innovation: Developed an improved BExM-based hydro-mechanical model that incorporates wall friction effects (based on swelling pressure and friction angle) and capillary water retention, validated against tested cases, to more accurately predict the evolution of bentonite intrusion distance over time.
36. Sequential sparse Bayesian learning of long-term thermomechanical performance of energy piles from limited multi-parameter monitoring data
Core Problem: Predicting the long-term thermomechanical performance of energy piles is challenging due to limited monitoring data, significant parameter uncertainty, and the high computational cost of fully coupled numerical analyses.
Key Innovation: Proposed a multi-objective sparse ensemble learning (MO-SEL) framework combining physics-informed load transfer functions with sequential sparse Bayesian learning to predict long-term pile performance, enabling accurate, uncertainty-aware predictions and reconstruction of full-field strain distributions from limited data.
37. Settlement Behavior of Bacterium-Embedded Soft Clay Sediment: Sedimentation and Low-Stress Consolidation Experiment
Core Problem: Lack of fundamental understanding of how bacteria influence the large-strain settlement and consolidation behavior of soft clay sediments, particularly in early diagenesis.
Key Innovation: Experimental demonstration that clay-bacteria interaction promotes aggregated fabric, leading to faster sedimentation, denser sediment, reduced compressibility, enhanced stiffness, and higher hydraulic conductivity in kaolin sediment, with a threshold stress identified for bacterial influence.
38. Quantitative video-based wave parameter estimation using a 3D-CNN and two-stage transfer learning framework
Core Problem: Conventional wave monitoring methods are costly, difficult to maintain, and limited in coverage, while existing video-based methods often lack robustness due to reliance on single-frame analysis or large labeled datasets.
Key Innovation: Development of a deep learning framework using a 3D-CNN with a two-stage transfer learning strategy (pre-training on simulated data, fine-tuning on small in-situ data) and POD-based denoising, achieving high accuracy in predicting significant wave height and peak period with reduced reliance on costly field measurements.
39. Experimental study and analysis of flow field characteristics of floating breakwaters with different cross-sectional shapes
Core Problem: Floating breakwaters are critical for mitigating marine environmental risks, but their shape significantly influences energy dissipation, requiring systematic investigation of wave attenuation and dynamic response for different designs.
Key Innovation: Systematic experimental and numerical investigation of wave attenuation efficiency and dynamic response of rectangular, trapezoidal, and triangular floating breakwaters, revealing that a 45° trapezoidal design effectively suppresses mooring force growth and that precise adjustment of inclination angle optimizes wave-attenuation performance.
40. Lithology identification based on drill-bit vibration signals and automated feature extraction for deep-sea drilling
Core Problem: Challenges in accurate lithology identification with limited measurement-while-drilling data, crucial for real-time decision-making in deep-sea drilling.
Key Innovation: Developed an intelligent lithology classification method using drill-bit vibration signals, integrating mechanistic analysis with Tsfresh-based automatic feature extraction and machine learning models (BiLSTM achieving 0.9651 accuracy), significantly improving performance under limited-feature conditions.
41. Graph Convolutional Networks Enable Fast Hemorrhagic Stroke Monitoring With Electrical Impedance Tomography
Core Problem: Fast and high-quality image reconstruction for hemorrhagic stroke monitoring with electrical impedance tomography (EIT) is computationally expensive using traditional nonlinear methods.
Key Innovation: A post-processing approach using graph convolutional networks (Graph U-net) significantly improved the image quality of linear difference reconstructions, achieving results comparable to or better than time-intensive nonlinear methods at negligible computational cost, and allowing 2D training for 3D processing.
42. AI-Driven Electrographic Seizure Classification and Seizure Onset Detection Using Image- and Time-Series-Based Approaches
Core Problem: Manually distinguishing between seizure and non-seizure events and precisely detecting seizure onset in intracranial electroencephalography (iEEG) recordings is highly time-consuming.
Key Innovation: Vision Transformers (ViTs) demonstrated superior performance in AI-driven electrographic seizure classification (97% accuracy) and seizure onset detection (1.4s median absolute error) from iEEG data, outperforming other image- and time-series-based AI models.
43. Deep learning–integrated multilayer thermal gradient sensing platform for real-time blood flow monitoring
Core Problem: Traditional blood flow monitoring methods are bulky or limited by blood vessel depth variability, hindering accurate and continuous assessment of cardiovascular health.
Key Innovation: Developed a soft electronic platform integrating multilayer thermal sensing with deep learning algorithms to simultaneously measure blood flow rate and vessel depth in real time, offering potential for personalized cardiovascular monitoring and early detection of hemodynamic events.
44. Probabilistic geological body modeling of point cloud based on IMMC-Geo and application
Core Problem: Accurate and cost-effective three-dimensional geological modeling under complex geological conditions with limited borehole data.
Key Innovation: Proposes IMMC-Geo framework integrating UAV imagery semantic segmentation (DeeplabV3+) for surface lithology extraction, virtual borehole generation from 3D point clouds, and a superimposed Markov chain with Monte Carlo simulation for stratigraphic inference, achieving ~80% consistency with conventional models and demonstrating robustness.
45. Experimental insights into CO<sub>2</sub> flow in fractured crystalline rock
Core Problem: The injection and flow of CO2 (non-wetting fluid) in fractured crystalline rock is a poorly understood process, critical for enhanced geothermal systems and in-situ carbon mineralization projects.
Key Innovation: A novel experimental method to evaluate the degree of non-wetting fluid (CO2) saturation in tight rock with nanometer scale pore sizes, using poromechanical and hydraulic measurements, revealing flow properties, breakthrough pressures, and relative permeabilities in fractured granite and rhyolite.
46. DHRFLUT: A scenario-controlled framework for decoupling runoff responses from land use transitions – a case study in China’s Wei River basin
Core Problem: Precisely quantifying the independent runoff effects of different land use transitions is challenging but critical for watershed management and understanding human-land interactions.
Key Innovation: Proposes DHRFLUT, a scenario-controlled framework that simulates individual land use transition scenarios and couples with the SWAT model to quantify runoff depth response coefficients (RDRC) at the sub-basin scale, providing a quantitative tool for assessing hydrological impacts of land planning.
47. A heterogeneous weighting strategy for leveraging Cross-Basin data enhances the Usability of deep learning hydrological models
Core Problem: Indiscriminate use of cross-basin data in training regional deep learning hydrological models can compromise performance at the local scale, leading to suboptimal predictions.
Key Innovation: Development of an enhanced network model with a novel heterogeneous weighting strategy that quantifies inter-basin influence to optimize deep learning model training for individual basins, significantly improving local and global hydrological predictions.
48. Modeling and characterizing errors in satellite and reanalysis precipitation estimates using a two-step decomposition procedure
Core Problem: Satellite and reanalysis quantitative precipitation estimates (QPEs) have distinct error structures that need to be characterized to improve their reliability for various applications, including hydrological modeling for geohazards.
Key Innovation: Application of a two-step decomposition approach to characterize and compare systematic and random errors in satellite (IMERG-Final, GSMaP-Gauge) and reanalysis (ERA5-Land) precipitation products, demonstrating the superiority of a multiplicative error model and providing region-specific diagnostic insights.
49. Numerical investigations of in-situ PRACLAY heater test: over-excavation, strain localisation and THM responses
Core Problem: Accurately analyzing the complex coupled thermo-hydro-mechanical (THM) processes and their impact on the stability and long-term performance of geological repositories for nuclear waste disposal, especially considering construction stages like over-excavation and the anisotropic behavior of host rocks like Boom Clay, is challenging.
Key Innovation: Implements an elasto-viscoplastic model with a hyperbolic Mohr-Coulomb yield surface to characterize Boom Clay, and innovatively uses air gap elements to represent over-excavation within a fully coupled THM framework. The model is verified and validated against in-situ PRACLAY Heater Test data, providing comprehensive insights into the effect of key parameters on THM behavior for nuclear waste disposal safety.
50. Moisture exposure detection in timber using NIR spectroscopy: A feasibility study for railroad condition assessment
Core Problem: Deterioration of timber railroad ties due to water exposure, and the need for a non-contact method to detect moisture transport from the ballast.
Key Innovation: Demonstrated the feasibility of using Near-InfraRed (NIR) hyperspectral imagery and Quadratic Discriminant Analysis (QDA) to detect moisture exposure in timber ties, achieving 89.7% testing accuracy. Identified prominent wavelengths around 1400 nm related to water absorption.
51. Ballast Aggregate Driven Nonlinear Environmental Stiffness in Heavy Haul Railway Fastenings: Novel Flexible Piezoelectric Sensor Monitoring
Core Problem: Conventional rigid sensors are inadequate for monitoring the structural integrity and dynamic performance of heavy-haul railway fastening systems, leading to challenges in assessing environmental stiffness and preventing derailments.
Key Innovation: Development and application of a novel, flexible, and conformable piezoelectric PVDF sensor material for accurately monitoring nonlinear environmental stiffness and dynamic interactions within heavy-haul railway fastening systems, enabling advanced track health monitoring and early-warning capabilities.
52. A Self-Levelling railway sleeper concept and its large-scale testing
Core Problem: Railway track transition zones suffer from abrupt stiffness changes, leading to permanent deformations and differential settlement (e.g., hanging sleepers), which are challenging and costly to mitigate with conventional methods.
Key Innovation: Design, development, and large-scale testing of modular self-levelling sleepers (SLS-G and SLS-HW) that effectively restore sleeper-ballast contact and improve load distribution, offering a low-disruption solution for mitigating track geometry degradation and reducing maintenance at transition zones.
53. A novel settlement calculation method and engineering application of strength composite piles
Core Problem: Existing settlement calculation methods for strength composite (SC) piles do not accurately reflect the stress diffusion and plastic zone evolution within the composite soil mass, especially in layered foundations.
Key Innovation: Proposal of a novel settlement calculation formula for strength composite (SC) piles based on multilayer theory, which accounts for the evolution of the plastic zone in surrounding soil and is applicable to both homogeneous and layered foundations, validated by static load tests and engineering applications.
54. More than ten years of hydration of an in situ large-scale sealing experiment (NSC), Meuse/Haute-Marne Underground Research Laboratory, France
Core Problem: Ensuring the long-term safety and performance of sealing cores in deep geological repositories for radioactive waste, specifically evaluating their in situ hydraulic permeability, swelling behavior, and gas performance over extended periods.
Key Innovation: Long-term (over ten years) in situ experimental assessment of a large-scale bentonite-sand sealing core, providing high-quality data on its resaturation, hydraulic, and gas performance, and evaluating the operational efficiency of various sensors for deep geological repository applications.
55. A novel multi-feature fusion Gaussian graph network for early anomaly detection of wind turbine
Core Problem: SCADA data from wind turbines is used for anomaly detection, but the complex interrelationships among variables are often overlooked, compromising accuracy.
Key Innovation: Proposal of a multi-feature fusion Gaussian graph network (MFGGN) that uses Gaussian radial basis function for graph construction and a graph attention network to adaptively aggregate spatiotemporal features from SCADA data, combined with the CUSUM method, achieving superior early anomaly detection performance for wind turbines.
56. Quantitative study on the impact of Stokes drift on marine oil spill transport and diffusion
Core Problem: Quantitatively assessing the specific role and contribution of Stokes drift to the transport and diffusion of marine oil spills, and determining the optimal forcing field for accurate oil spill modeling.
Key Innovation: Designing sensitivity experiments using various forcing field data to drive an oil spill model, demonstrating that coupled current, wind, and model-derived Stokes drift achieve optimal simulation accuracy, and quantitatively showing Stokes drift's significant contribution (over half of wind-induced drift when wind speeds exceed 4.5 m/s) with a lag effect relative to wind speed changes.
57. Wave-induced transport of non-buoyant microplastic particles: Phase-resolved experiments and excess-Shields scaling
Core Problem: Understanding the bed-load dynamics and phase-resolved motion of non-buoyant microplastic particles under wave forcing in coastal environments.
Key Innovation: Phase-resolved experiments establishing a link between particle mobilization and the modified Shields parameter, identifying a residual onshore drift dependent on particle characteristics and wave steepness, and revealing a non-linear dependence between particle mobility and applied force.
58. GVCCS: a dataset for contrail identification and tracking on visible whole sky camera sequences
Core Problem: Existing contrail datasets lack temporal tracking and attribution to source flights, limiting the validation and calibration of physics-based models for understanding aviation's non-CO2 climate impact.
Key Innovation: Presents the Ground Visible Camera Contrail Sequences (GVCCS) dataset, an open dataset of 122 video sequences with individually labeled and tracked contrails, including flight identifiers. Also proposes a unified deep learning framework for contrail analysis using a panoptic segmentation model.
59. Building a dataset of offshore oil and gas extraction platforms from satellite data (2017–2023)
Core Problem: Existing databases of offshore oil and gas platforms (OOGPs) are often incomplete or outdated, hindering informed decision-making and evaluation of environmental impacts.
Key Innovation: Constructed a new database of 5,358 OOGPs for six major offshore basins (2017-2023) using Sentinel-1 SAR satellite data and Google Earth Engine, achieving 98% extraction accuracy and providing enhanced spatiotemporal coverage compared to existing databases.
60. Open-source AI tool beats giant LLMs in literature reviews — and gets citations right
Core Problem: Existing large language models (LLMs) for literature reviews can be costly, lack transparency, and struggle with accurate citation generation.
Key Innovation: Developed an open-source, deployable AI tool that outperforms large LLMs in literature reviews, offering cost-effectiveness, transparency, and accurate citation handling.
61. As the world warms, freezing rain shifts to U.S. South
Core Problem: Understanding the geographical shifts in weather phenomena, specifically freezing rain, as a consequence of global warming.
Key Innovation: Identifying and documenting a shift in freezing rain occurrences towards the U.S. South due to global warming.
62. Ozone photochemistry in fresh biomass burning smoke over the United States
Core Problem: Understanding the complex ozone photochemistry and the influence of volatile organic compounds (VOCs), nitrogen oxides, and nitrous acid in fresh biomass burning plumes, and identifying gaps in current chemical transport models.
Key Innovation: Analysis of aircraft campaign data revealing variable, highly elevated hydroxyl radical concentrations, fire-to-fire variability in VOCs oxidation and ozone/PAN production, and identification of critical pathways to guide future model development (e.g., MCM overestimates PAN, GEOS-Chem underperforms due to incomplete VOC representation).
63. Editorial: Advances and new methods in reservoirs quantitative characterization using seismic data
Core Problem: The practical challenges in seismic-driven reservoir characterization due to band-limited measurements, imperfect illumination, and sparse calibration, while still needing to produce models that can guide real decisions.
Key Innovation: An editorial highlighting a collection of concrete methods, inversion strategies, signal-description tools, and field-based demonstrations for quantitative reservoir characterization using seismic data, aiming to provide a working reference for practitioners.
64. An advanced petri-net modelling approach for risk asset management of reinforced concrete ageing transportation infrastructure under climate change effects
Core Problem: Managing the risk of ageing reinforced concrete transportation infrastructure under the accelerating effects of climate change, specifically addressing deterioration mechanisms, inspection policies, and maintenance strategies.
Key Innovation: Proposes an advanced modular Petri Net (PN) framework for risk-based asset management, integrating probabilistic degradation processes with climate-dependent parameters, multivariate statistical formulation for interdependencies, and demonstrating its application to an RC bridge using UKCP18 climate projections to highlight accelerated corrosion and the need for adaptive strategies.
65. Fire resilience in station-city integrated spaces: A multi-stage assessment and optimization framework incorporating spatial and behavioral interactions
Core Problem: Major fire safety challenges in complex, high-pedestrian-density Station-City Integrated Spaces (SCIS) due to urbanization and expanding transport hubs.
Key Innovation: A multi-stage resilience assessment and optimization framework that integrates spatial and behavioral dimensions to improve fire safety management in SCIS, identifying key resilience factors and guiding targeted strategies.
66. Uncertainty informed calibration of thermal-hydraulic models for nuclear reactor via integrated neural network and optimization algorithm framework
Core Problem: Calibrating thermal-hydraulic models for nuclear reactors is difficult due to insufficient data, complex model structures, numerous parameters, and the inability of non-probabilistic methods to account for model form uncertainty.
Key Innovation: Proposes a novel uncertainty-informed calibration framework based on non-probabilistic interval theory, integrating artificial neural networks, model uncertainty evaluation, double-loop nested sampling, and optimization algorithms to obtain accurate input parameter intervals with high computational efficiency, even with limited data.
67. Data-driven copula Bayesian network for risk analysis of lithium-ion battery accidents
Core Problem: Risk analysis of lithium-ion battery thermal runaway accidents is crucial, but traditional methods may rely heavily on expert experience, and capturing nonlinear features and model uncertainty is challenging.
Key Innovation: Proposes a data-driven copula Bayesian network (DCBN) model for risk analysis of LIB accidents, using copula functions to capture nonlinear correlations and a data-driven method to depict causal relationships between risk-influencing factors, validated with aviation LIB data.
68. Gravity maps of the African continental crustal and mantle structure
Core Problem: Knowledge of Africa's lithospheric structure is limited by sparsely distributed seismic surveys, hindering continent-wide analysis and detailed understanding of its complex tectonic and geological history.
Key Innovation: Compiles a suite of high-resolution gravity maps (free-air, Bouguer, crust-stripped, mantle, lithosphere-stripped, sub-lithospheric mantle) on a 5′ × 5′ grid, integrating satellite and terrestrial gravity observations with advanced corrections. These maps enable detailed interpretation of Africa's lithospheric architecture, revealing insights into crustal thickness variations, thermal signatures of active rifts and stable cratons, and geodynamic evolution, supporting future geophysical and resource exploration efforts.
69. The impact of climate change on vertical hydrological connectivity of wetlands
Core Problem: Vertical hydrological connectivity of wetlands (VHCW) is significantly altered by climate change and human activities, leading to ecological degradation risks and increased uncertainty in VHCW function in Western Jilin Province.
Key Innovation: Developed and validated a SWAT hydrological model with a wetland module to predict spatiotemporal variations of VHCW during historical and future climate scenarios. Found that VHCW will mainly manifest as wetland surface water recharge to groundwater, with significant differentiated impacts of future emission pathways.
70. Comprehensive framework for agricultural water management in data-scarce regions: Integration of hydrological models and remotely sensed crop type data
Core Problem: Effective agricultural water management is challenging in data-scarce regions, requiring integrated approaches for hydrological modeling and crop type mapping to accurately assess water demand and balance.
Key Innovation: A comprehensive framework integrating the WRF-Hydro model for daily discharge simulation and an ensemble of machine learning algorithms with MODIS NDVI data for crop type prediction, to estimate irrigation water requirements and assess water balance in data-scarce regions.
71. From molecular mechanisms to remote sensing retrievals: a physically consistent framework for fluorescent dissolved organic matter
Core Problem: Most remote sensing retrievals for fluorescent dissolved organic matter (fDOM) rely on empirical statistics, lacking physical interpretation of fluorescence mechanisms and radiative transfer, which limits their applicability in optically complex waters.
Key Innovation: Developed a molecular mechanism-based quantitative inversion method for fDOM remote sensing, incorporating a molecular two-level transition model, radiative transfer approximations, and inner-filter correction, validated with field measurements, providing a physically grounded approach for optically complex aquatic environments.
72. Four decades of terrestrial dissolved organic matter in lakes across the Yangtze River Delta: Spatiotemporal dynamics and driving factors
Core Problem: Quantifying the fluorescence intensity and spatiotemporal patterns of terrestrial dissolved organic matter (tDOM) in lakes is a persistent challenge for traditional monitoring, creating a critical gap in regional carbon budget estimations and understanding large-scale ecosystem effects.
Key Innovation: Developed a novel remote sensing algorithm to retrieve tDOM based on the absorption characteristics of colored dissolved organic matter (CDOM), applied it to assess four decades of spatiotemporal dynamics in Yangtze River Delta lakes, and used a random forest model to identify temperature, wind speed, and anthropogenic pressure as key drivers.
73. Crack initiation and propagation mechanism of borehole surrounding rock subjected to cyclic thermal loading: insights from theoretical solution and DEM simulation
Core Problem: Understanding the variation of temperature and stress fields and the crack initiation and propagation mechanism in borehole surrounding rock during cooling processes for geothermal reservoir construction.
Key Innovation: Derivation of unsteady temperature and stress fields, proposal of crack initiation and propagation criteria, and investigation of thermal crack laws using the Discrete Element Method (DEM), revealing factors influencing cracking and providing theoretical reference for EGS reservoir construction.
74. Soil compaction control by monitoring compacted states based on soil stiffness indices
Core Problem: Accurately estimating and controlling the compacted water content and dry density of soil during compaction is crucial for construction quality, but current methods based on soil stiffness indices (SSIs) need better generalization and practical application guidelines.
Key Innovation: Generalization of empirical equations for soil stiffness indices (SSIs) into a normalized form and proposal of a simplified method for soil compaction control, enabling the estimation and maintenance of compacted water content within a target range by monitoring field SSIs between defined thresholds.