TerraMosaic Daily Digest: Feb 14, 2026
Daily Summary
Across 71 selected papers from 631 newly analyzed studies, the strongest signal is fluid-solid coupling in soils, seabeds, and engineered systems. Core contributions quantify how fines content, void ratio, and intergranular state govern internal erosion, how bentonite treatment pathways shift liquefaction resistance, and how seepage and bore-like wave loading drive stress and sediment mobilization around permeable slopes and buried infrastructure.
In parallel, risk analysis is becoming more mechanism-aware and decision-ready. Interpretable flood-governance models separate hydrologic forcing from spatial heterogeneity, hypergraph and interdependent-network frameworks resolve cascade pathways in critical infrastructure, and resilience metrics are linked to targeted fortification and maintenance actions. Observation-constrained studies, including Sentinel-1 estimates of coastal vertical land motion and broadband magnetotelluric imaging beneath Sabancaya, tighten exposure and process constraints needed for robust hazard assessment.
Key Trends
- Soil–water instabilities are being parameterized at the grain scale: Internal erosion, liquefaction, and seepage-driven deformation are linked to fabric descriptors (e.g., intergranular state) and to intervention pathways (how fines or bentonite are introduced), not just to bulk density or fines content.
- Coastal and offshore hazards are framed as porous-media mechanics problems: Wave run-up, pore-pressure evolution, and stress within permeable slopes/seabeds are solved together, improving how we translate hydrodynamics into slope or foundation vulnerability.
- Cascading-risk analytics now encode infrastructure heterogeneity explicitly: Flood and rainstorm cascades are modeled with heterogeneous state transitions and hypergraph/interdependent-network structure, supporting targeted fortification rather than uniform hardening.
- Explainable AI is being used to specify interventions, not only to predict hazards: Interpretable models decouple hydrologic process from spatial heterogeneity and map outputs to governance units that can be acted on (zoning, drainage upgrades, prioritized maintenance).
- Deformation and subsurface imaging are tightening constraints on exposure and drivers: Sentinel‑1 InSAR and magnetotellurics are used to quantify vertical land motion, identify subsurface fluid pathways, and connect those measurements to population and asset exposure.
Selected Papers
This digest features 71 selected papers from 631 newly analyzed papers across multiple journals. Each paper has been evaluated for its relevance to landslide and broader geohazard research and includes links to the original publications.
1. Synergistic effect of fines content and void ratio on internal erosion: An intergranular state perspective
Core Problem: The synergistic effect of fines content and void ratio on internal erosion, a major cause of hydraulic structure failure, is not fully understood, particularly how soil geometric characteristics influence hydraulic and structural responses during this process.
Key Innovation: Systematically investigated the synergistic effect of fines content and void ratio on internal erosion in gap-graded mixtures, revealing three characteristic erosion stages, proposing a robust empirical model for critical hydraulic gradients, and introducing the intergranular void ratio (e_s) as a unified descriptor linking microstructural configuration to hydraulic instability and seepage-induced deformation, thereby improving the assessment of skeleton stability.
2. Magma Storage Below Sabancaya Volcano (Southern Peru) Imaged by Broadband Magnetotellurics
Core Problem: Understanding the magmatic plumbing system and processes driving large-scale unrest (deformation, seismicity) at active volcanoes like Sabancaya is essential for hazard assessment.
Key Innovation: Uses a pilot magnetotelluric survey to create a 3D electrical resistivity model, revealing prominent conductors interpreted as magma reservoirs and a highly conductive body in the Huambo-Cabanaconde fault zone attributed to ultra-saline brines. This provides a model of crustal fluid distribution and connects deep magma reservoirs to shallow chambers and fault systems.
3. Vertical Land Motion and Human Exposure Across India's Coastal Regions
Core Problem: The role of vertical land motion (VLM), particularly land subsidence, in shaping coastal exposure and increasing flood risks in India remains poorly quantified.
Key Innovation: Presents the first comprehensive assessment of VLM across India's coastal zone using 8 years of Sentinel-1 InSAR data, revealing widespread subsidence (up to 20 mm/yr in major deltas) affecting over 8.5 million residents, and highlighting its significant challenge for long-term coastal land-use planning.
4. Bore-like waves propagation and impact on a permeable slope over a dry bed: Impact loads, stress distribution, and run-up
Core Problem: A comprehensive understanding of bore-like wave propagation and impact on permeable slopes, including detailed hydrodynamic responses, impact forces, and stress distributions within the porous media, is needed for accurate hazard assessment and design.
Key Innovation: This study uses a modified Darcy–Brinkman–Biot model to examine bore-like wave run-up on permeable slopes, quantifying the influence of slope angle, permeability, and porosity on flow structures, intrusion volume, run-up height, impact forces, and von Mises stress distributions, providing quantitative insights into wave-structure interactions in porous media.
5. Revealing the Dynamics and Multidimensional Resilience of Rainstorm-Flood Cascade Disasters in Mountain Valley Cities: An Interpretable Machine Learning Case Study from Southwestern China
Core Problem: The dynamic evolution and multidimensional resilience effects of cascade disasters triggered by extreme rainstorm-flood events in mountainous valley cities remain unclear, limiting effective disaster management.
Key Innovation: A framework combining complex networks, t-SNE dimensionality reduction, and K-Means clustering to assess disaster resilience to rainstorm-flood cascade events, revealing shifts in vulnerability and non-stationary resilience patterns in different zones for improved disaster prevention policies.
6. Exploring cascade propagation mechanisms of flood risk in dynamic heterogeneous infrastructure networks: a systematic framework with a heterogeneity perspective
Core Problem: Predicting and controlling flood damage risks in infrastructure networks is difficult due to increasing flood frequency/intensity, and traditional epidemic models often rely on homogeneous assumptions, failing to capture the coupling effects of dynamic heterogeneous elements on risk propagation.
Key Innovation: Proposes a coupled analysis framework integrating the Infrastructure Hetero-Functional Graph with the Susceptible–Degraded–Failure–Recovered model to comprehensively incorporate heterogeneous infrastructure parameters under flood impacts, establish dynamic state transition probabilities, and evaluate risk control strategies.
7. An explainable artificial intelligence approach for synergistic flood risk governance by decoupling hydrological process and spatial heterogeneity
Core Problem: While AI has advanced urban flood risk prediction, its 'black-box' nature and limited interpretability hinder the translation of model outputs into actionable governance strategies, particularly in capturing multi-factor interactions and supporting direct regulatory decision-making.
Key Innovation: Developed a governance-oriented explainable AI (XAI) approach that decouples hydrological processes and spatial heterogeneity to analyze flood mechanisms, establishing a transmission pathway from macro-level risk recognition to micro-level intervention design, and delineating 8 distinct governance units with strong alignment to actual flood records, thereby transforming predictive capability into practical decision intelligence for flood governance.
8. A three-dimensional arbitrary grid material point method for large deformation problems with geometrically complex boundaries
Core Problem: The Material Point Method (MPM) faces bottlenecks in modeling large deformation problems with geometrically complex boundaries, especially in three dimensions, due to computational costs and boundary challenges.
Key Innovation: A three-dimensional arbitrary grid material point method (3D-AGMPM) that uses polyhedral grid cells, 3D-Wachspress basis functions, and an improved hash-cell based particle localization algorithm to efficiently and robustly simulate large deformation problems with complex geometric boundaries, demonstrated with practical case studies like the Wangjiayan landslide.
9. Influence of the type and method of bentonite addition on the liquefaction resistance of a standard clean sand
Core Problem: Conventional ground improvement techniques for liquefiable soils are limited by volumetric changes, and the effectiveness of alternative plastic nanoparticles like bentonite in enhancing liquefaction resistance, particularly concerning the method of addition (dry vs. suspension), is not fully understood.
Key Innovation: Systematically evaluates the influence of bentonite type and addition method (dry mixing vs. permeation) on the liquefaction resistance of sand, demonstrating that permeation with bentonite suspensions significantly enhances cyclic resistance, while dry mixing can reduce it, providing insights for effective liquefaction mitigation.
10. From Collision to Subduction: Thermal‐Kinematic Inversions Constrain Plate Boundary Structure and Dip‐Slip Activity in Southwestern New Zealand
Core Problem: Understanding the three-dimensional fault system geometry and kinematics, and late Cenozoic dip-slip activity of the plate boundary in southwest New Zealand, is crucial for assessing contemporary seismic hazard.
Key Innovation: Inverts low-temperature thermochronological data to quantitatively parameterize crustal-scale geometry and late Cenozoic dip-slip activity of the plate boundary system. This resolves shortening rates and elucidates the transition from oblique continental collision to strain-partitioned subduction, providing constraints on a structural feature that may have arrested major prehistoric earthquakes.
11. Effect of seepage flow on sediment incipient motion around buried pipelines
Core Problem: Previous numerical studies on sediment incipient motion around pipelines often simplify the seabed as impermeable, underestimating the effects of seepage flow and thus the risk of sediment entrainment and localized scouring.
Key Innovation: This study integrates a porous seabed model with a turbulence model to simulate seabed responses around buried pipelines, demonstrating that pore water pressure accumulation significantly increases seepage, promoting sediment incipient motion. It quantifies the asymmetric influence of pipelines on sediment entrainment and shows how soil permeability and shear modulus affect this process.
12. Reliability design of reshaping berm breakwaters based on a Bayesian modification
Core Problem: Existing experimental formulas for predicting berm recession in reshaping berm breakwaters are often not well-suited for probabilistic reliability design, and there's a lack of reliability-based design tools that incorporate the influences of sequential storms.
Key Innovation: This study develops probabilistic formulas for the reliability design of reshaping berm breakwaters by homogenizing data from rebuilt and cumulative tests to incorporate sequential storm effects. It applies Bayesian Linear Regression (BLR) to derive a comprehensive prediction formula and a simplified design formula, providing a novel practical approach for reliable design that demonstrates high accuracy and robustness.
13. Integrating FELA with EPR MOGA-XL and XGBoost for assessing uplift capacity of buried pipelines in dense sand
Core Problem: Quantifying the uplift behavior and improving the uplift resistance of buried subsea pipelines, especially finned pipelines in dense sand where stress-dependent dilatancy is critical, remains a challenge, and efficient predictive tools for design are needed.
Key Innovation: This study proposes a hybrid framework integrating Finite Element Limit Analysis (FELA) with machine learning (XGBoost for prediction, EPR Moga-XL for interpretable equations) to assess the uplift capacity of finned pipelines in dense sand. It quantifies that fins can enhance uplift resistance by up to 30% and identifies embedment ratio and particle crushing strength as dominant factors, providing both mechanistic insight and efficient predictive tools for design.
14. Robustness of disaster response systems: Hypergraph modelling, key local structures, and fortification methods
Core Problem: Disaster response systems (DRS) operate in uncertain environments and need to withstand disruptions, but existing robustness metrics from social network analysis are not always direct or well-established for maintaining structural integrity.
Key Innovation: Constructed faithful hypergraph representations of DRS to reveal key local structures for structural integrity, established comprehensive robustness metrics, and developed efficient fortification methods to maximize total robustness gain, outperforming alternative strategies.
15. Robustness of spatial interdependent networks under extreme geographically localized attacks
Core Problem: Current approaches for identifying critical locations of extreme geographically localized attacks (EGLA) in spatial interdependent networks primarily focus on single-layer networks, ignoring interlayer interdependencies and leading to erroneous robustness estimates.
Key Innovation: An optimization model for identifying critical locations in spatial interdependent networks and a combinatorial region search simulated annealing algorithm (CRS-SA) that effectively considers cascading failures and provides insights for enhancing network security against EGLA.
16. Resilience Assessment of Electrical Distribution Networks Subject to Wind and Self-Induced Fire Hazards
Core Problem: Aging electrical distribution networks are increasingly vulnerable to hazards like wind and fire, leading to widespread failures and prolonged outages, requiring better resilience assessment and repair strategies.
Key Innovation: A simulation-based framework to evaluate the resilience of an electrical distribution system under combined urban fire and wind hazards, modeling probabilistic pole/conductor failures and time-stepped restoration, providing insights for optimizing repair strategies and workforce deployment.
17. A Y-shaped artificial boundary and earthquake input scheme for step-shaped layered site and soil-structure interaction analyses
Core Problem: Conventional methods for seismic soil-structure interaction (SSI) are limited to horizontally layered half-space sites, making it challenging to accurately evaluate site seismic responses and implement earthquake input for non-horizontally layered step-shaped sites.
Key Innovation: A high-precision Y-shaped artificial boundary and earthquake input scheme for step-shaped layered sites and SSI analyses, which uses two zigzag-paraxial combined boundaries to simulate radiation damping and transforms 1D site responses into equivalent nodal forces for earthquake input, enabling reliable SSI evaluation in complex step-shaped site conditions.
18. Physics‐Based Versus AI Weather Prediction Models: A Comparative Performance Assessment of Atmospheric River Prediction
Core Problem: The reliability of emerging ML weather prediction models in accurately forecasting high-impact weather events like Atmospheric Rivers (ARs), especially compared to traditional physics-based models, and the inadequacy of standard metrics for assessing their performance for such events.
Key Innovation: A comparative performance assessment of five state-of-the-art ML models against a high-performing physics-based model (HRES) for U.S. West Coast AR prediction, revealing that HRES has superior AR detection skill for the first four forecast days, and highlighting the need for phenomenon-specific metrics for ML-based numerical weather prediction model assessment.
19. A 1D-CNN deep learning framework for seismic collapse prediction of jacket offshore platforms with Bayesian neural architecture search
Core Problem: Accurately predicting seismic limit state exceedance (collapse) in Jacket Type Offshore Platforms using earthquake accelerogram time series, and systematically optimizing the deep learning model architecture and hyperparameters for this task.
Key Innovation: Developed a 1D-CNN deep learning framework with Bayesian optimization for neural architecture search to predict seismic collapse of offshore platforms, demonstrating its capability for computing collapse fragility curves from earthquake time series.
20. Forecasting ecosystem service values in South China’s fragile mountain regions: a multi-scenario analysis with adaptive land use valuation
Core Problem: Reconciling sustainable development and ecological conservation in economically underdeveloped and ecologically fragile mountain regions of South China, by understanding how ecosystem service values respond to different land use scenarios.
Key Innovation: Integrated historical land use data and PLUS modeling with adaptive valuation coefficients to forecast ecosystem service values under multi-scenarios (NGS, FPS, EPS) in a fragile mountain region, demonstrating that the NGS achieves the greatest ESV increase while the EPS offers balanced development through multidimensional service enhancements, underscoring the need for ecological safeguards in land use planning.
21. Application of a Novel Anisotropy Modulation Function to Evaluate Strength Behavior of Layered Rock
Core Problem: Accurately capturing the anisotropic mechanical behavior of layered rock, particularly under true triaxial stress, which existing failure criteria may not fully address.
Key Innovation: A novel anisotropy modulation function incorporated into a modified failure criterion that accurately predicts the strength behavior of layered rocks under both conventional and true triaxial stress states, considering bedding inclination and intermediate principal stress effects.
22. High-frequency observation reveals the impacts of suspended sediment and hysteresis effect on water quality in soil erosion areas
Core Problem: The impacts, relationships, and driving factors of suspended sediment on water quality, particularly the hysteresis patterns during rainfall events in soil erosion areas, remain unclear due to limited high-frequency observations.
Key Innovation: Utilized six years of high-frequency observations to reveal the impacts and hysteresis patterns of suspended sediment on water quality in soil erosion areas, demonstrating increased phosphorus variability during heavy rainfall, intensified nitrogen hysteresis, and distinct counterclockwise flushing hysteresis for water quality indicators, highlighting the critical role of suspended sediment for monitoring and early warning of water pollution.
23. Experimental Study on the Accumulation Law of Creep Deformation in Rockfill Materials under Multistage Loading
Core Problem: The complex coupling effect between stress loading and creep behavior in rockfill materials, particularly the accumulation law of creep deformations under different multistage loading conditions (regular vs. small increments), is not fully understood, hindering accurate long-term deformation prediction for structures like high rockfill dams.
Key Innovation: Experimental investigation revealing distinct creep accumulation patterns under different loading conditions, leading to the establishment of a new accumulation law for rockfill material creep that defines current creep strain as equal to hardening creep strain, offering a universally applicable framework for predicting long-term deformation under complex loading.
24. Molecular models identify grain boundaries as accumulating dislocations and promoting creep in rock salt
Core Problem: Traditional models inadequately explain the coevolution of dislocation dynamics and pore networks under sustained loading in rock salt, hindering prediction of its creep behavior and long-term stability relevant to salt tectonics and cavern integrity.
Key Innovation: Integrated multi-scale techniques (X-ray microcomputed tomography, EBSD, molecular dynamics) to reveal that grain boundaries in rock salt play a dual role in accumulating dislocations and dissipating energy, which underpins its remarkable ductility and provides key physics-based insights for predicting the long-term stability of geological and engineered salt systems.
25. Quantifying contribution of soil spring in multi-spring models for laterally loaded monopiles
Core Problem: The commonly used API P-y method for designing laterally loaded monopiles, especially rigid ones with low slenderness ratios, fails to accurately capture three-dimensional soil-monopile interaction effects, leading to an incomplete understanding of individual soil spring contributions in multi-spring models.
Key Innovation: This study systematically quantifies the contribution of each soil spring in multi-spring models for laterally loaded monopiles using Abaqus, revealing that for low slenderness ratios (L em/D < 6), tip and shaft moment/shear resistances make non-negligible contributions. It introduces a slenderness threshold to guide efficient and reliable monopile design, improving the accuracy of soil-monopile interaction modeling.
26. Experimental study on the adhesive characteristics between clayey soil and metal in a slurry immersion environment
Core Problem: Existing adhesion mechanisms for clayey soil and metal are not applicable to slurry shield tunneling environments, leading to challenges in assessing and mitigating cutterhead clogging.
Key Innovation: Demonstrated how slurry immersion alters adhesive stress, introduced the “Adhesion-Active Layer” concept, and developed an “Adhesion Evolution Path Model” to predict clogging potential in slurry shields.
27. Fully Coupled Plastic-Softening Porosity-Based THM Model of Supercharged Zone Evolution around the Wellbore with Dynamic Mud Cake Growth
Core Problem: Unquantified synergistic effects of Thermo-Hydro-Mechanical (THM) coupling, plastic deformation with strength softening, and dynamic mud cake growth on the evolution of the supercharged zone around a wellbore, hindering accurate prediction of formation damage and pore pressure.
Key Innovation: Introduction of a novel, fully coupled Plastic-Softening-Porosity-Based (PSPB) THM model that integrates these complex factors, revealing a three-stage temporal evolution of pore pressure response and establishing a definitive hierarchy of influence on pore pressure localization (in-situ stress anisotropy and wellbore pressure being most critical).
28. Desiccation Cracking Behavior of Carbon Microfiber Reinforced Bentonite Clay for Deep Geological Repositories
Core Problem: Desiccation cracking in bentonite clay, used as an engineered barrier in deep geological repositories, can compromise its integrity by creating pathways for water, leading to potential nuclear waste leakage.
Key Innovation: Demonstrates that carbon microfiber reinforcement significantly reduces macrocracks and induces microcracks in bentonite clay during desiccation, improving its integrity for deep geological repositories, with 1.0% reinforcement showing optimal performance.
29. The Impact of Weather and Environmental Factors on Power Outage Vulnerability
Core Problem: Weather-related power outages are becoming increasingly frequent and severe, posing significant challenges for grid reliability and public safety, necessitating quantification of town-level outage risks and identification of key environmental drivers of vulnerability.
Key Innovation: A data-driven approach that quantifies town-level power outage risks using return period thresholds (estimated from GEV distributions) and identifies key environmental drivers (dense canopy cover, high elevation, stronger winds) through a Generalized Linear Model, contributing to effective emergency preparedness and resilience planning.
30. Deep foundation pit excavation risk identification and mitigation using knowledge graph integrated with domain knowledge and monitoring data
Core Problem: Risk identification and mitigation (RIM) in deep foundation pit excavation is largely manual, relying on subjective judgment and limited knowledge application to sensor data, creating a bottleneck in automation.
Key Innovation: A knowledge-driven logical framework integrating monitoring data with RIM processes via a risk management knowledge graph (RMKG), enabling automated analysis of sensor data, knowledge extraction from reports, and automatic formulation of response measures for enhanced safety resilience.
31. A multi-objective simulation-optimisation model for managing saltwater intrusion in a coastal aquifer of the Pacific Island of Vanuatu
Core Problem: Managing saltwater intrusion (SWI) in coastal aquifers, particularly in data-limited regions like Pacific Islands, requires robust predictive and optimization models that can handle complex hydrological processes and provide actionable policies.
Key Innovation: Developed and validated a novel multi-objective simulation-optimization (S-O) management model, integrating numerical (SEAWAT) and ensemble machine learning surrogate models with a genetic algorithm (CEMOGA), demonstrating high predictive accuracy and global applicability for SWI management, especially with limited datasets.
32. Modeling runoff with incomplete data: a comparison of hydrological, deep learning, and hybrid approaches
Core Problem: There is a limited understanding of runoff model performance and sensitivity under varying degrees and types of input data scarcity, which affects effective water resources management and hazard prediction.
Key Innovation: A systematic evaluation comparing physical process-based (Xinanjiang), data-driven (LSTM), and hybrid runoff models under different data scarcity scenarios, demonstrating the reliability of process-based models under data-scarce conditions and the benefits of hybrid approaches, informing model selection for practical hydrological applications.
33. Monitoring of supporting soil during construction of 330-m supertall building on spread foundation in Tokyo
Core Problem: The unprecedented scale and complex construction process of a 330-m supertall building on a spread foundation in Tokyo could lead to complex and difficult-to-predict behaviors in the supporting soil, necessitating crucial monitoring to ensure construction quality.
Key Innovation: Detailed monitoring of vertical displacement and stiffness changes (shear wave velocity) of the supporting soil during the construction of the Mori JP Tower, revealing specific rebound and settlement patterns, the influence of groundwater level, and confining-pressure dependency of shear wave velocity, which generally align with previous reports and contribute to understanding soil behavior under extreme loading.
34. Dynamic response of a floating offshore wind turbine under multi-stage typhoon conditions: A case study of super typhoon In-Fa
Core Problem: Existing Floating Offshore Wind Turbines (FOWTs) lack sufficient resilience to extreme weather events like typhoons, with conventional parked strategies failing to ensure mooring safety and manage complex structural loads.
Key Innovation: This study investigates a novel low-rotor-speed control strategy for FOWTs under multi-stage typhoon conditions, identifying the Front Eye Wall stage as dominant. It reveals a trade-off where low-rotor-speed increases mean loads but mitigates peak tower-base shear, while highlighting that both conventional and novel strategies can compromise mooring safety due to large platform surge excursions, underscoring the need for enhanced mooring capacity.
35. Comprehensive evaluation and zoning study of potential geothermal water resources in Qinghai Province, northeastern margin of the Tibetan Plateau
Core Problem: Geothermal resources in Qinghai Province are underexplored, spatially uneven, and carry high exploration risks due to a lack of systematic exploration in many sedimentary basins and structural fault zones.
Key Innovation: Developed a novel semi-quantitative methodology integrating basin heat storage and tectonic heat control theories, establishing a multidimensional evaluation index system based on 'source-pathway-cap-reservoir-flow' genetic elements. This method, combining AHP and PCA, quantitatively assessed and zoned geothermal potential, delineating 32 prospective areas and prioritizing 10 target areas for early-stage exploration.
36. Natural and anthropogenic radionuclides in bottom sediments of flooded delta lakes of the pechora river (Arctic ocean Basin)
Core Problem: Understanding the activity and distribution of natural and anthropogenic radionuclides in bottom sediments of flooded delta lakes, and how the complex hydrological regime and permafrost-related thermoabrasion influence pollutant accumulation and the reliability of chronological indicators.
Key Innovation: Quantified radionuclide activities in Pechora River delta lake sediments, found them commensurate with global averages, and demonstrated that the complex hydrological regime and thermoabrasion in permafrost areas significantly impact sedimentation, pollutant accumulation, and render vertical radionuclide distribution an unreliable chronological indicator.
37. Land use impacts on soil organic carbon stratification and sequestration in a karst region of Guizhou, China
Core Problem: Understanding how land-use changes influence the vertical and temporal dynamics of soil organic carbon and its labile fractions, and consequently soil carbon storage and ecological restoration processes, in a vulnerable karst region.
Key Innovation: Quantified SOC, ROOC, and WSOC dynamics across depths and land-use types in a karst region, identified shrubland as having the highest surface carbon sequestration efficiency, and advocated for stratified land management (shrubland promotion, reduced tillage) to enhance carbon retention in these fragile ecosystems.
38. Investigation on the Fracture Characteristics and Mechanism of Hard Rock by Microwave Irradiation in Underwater Environment
Core Problem: Lack of understanding regarding the fracture characteristics and mechanism of hard rock when subjected to microwave irradiation in underwater environments, which is crucial for safe and efficient deep underground or sub-sea resource development.
Key Innovation: Experimental and numerical investigation using a self-made sealed microwave antenna to study microwave cracking processes in three hard rocks underwater, revealing distinct damage patterns (spalling vs. cracking), acoustic emission characteristics, and significant improvements in cutting depth after microwave pretreatment, along with insights into temperature/stress distributions.
39. Resilience assessment of urban mobility flow networks from different scales: A case study in shenzhen
Core Problem: Urban mobility flow networks are vulnerable to disruptions, and there's a need for a multi-scale framework to assess their resilience to understand impacts and recovery processes across different spatial scales.
Key Innovation: Presented a multi-scale framework (macro, meso, micro) to assess urban mobility flow network resilience, identifying the influence of high-degree nodes/edges and centrality metrics on network performance and recovery dynamics under disruptions.
40. Data-Analytics Identification of Heterogeneous Component Criticality in Coupled Transportation-Power Systems Considering Bidirectional Cascading Effects
Core Problem: Existing critical component identification methods for coupled transportation-power systems (CTPS) do not adequately consider bi-directional cascading effects, leading to an incomplete understanding of heterogeneous component criticalities.
Key Innovation: Proposes a data-analytics framework using a co-simulation method to generate massive cascading effect data and a SALSA algorithm-based data-analytics method to evaluate heterogeneous component criticalities, considering propagation path risk for prevention and risk management.
41. Risk-Informed Opportunistic Operation and Maintenance for Floating Offshore Wind Turbine Towers Subjected to Corrosion Fatigue
Core Problem: Floating Offshore Wind Turbines (FOWTs) face structural risks from corrosion-fatigue (CF) deterioration, requiring a comprehensive framework to evaluate and mitigate CF risk and optimize operational costs and power production.
Key Innovation: Develops a comprehensive framework integrating multi-physics modeling, reliability assessment, and a Genetic Algorithm-driven co-scheduled opportunistic operation and maintenance (OOM) framework to optimize turbine control strategies and maintenance actions, mitigating CF risk and extending service life.
42. Gait-Aware quadruped 3D mapping in challenging environments with complex terrain
Core Problem: Quadruped SLAM systems suffer from severe motion distortions and accumulated drift in complex environments due to frequent gait transitions, unstable ground contacts, and abrupt body motions.
Key Innovation: Proposes a novel gait-aware quadruped 3D mapping method that incorporates gait-aware constraints and confidence-aware updates (GA-ESKF, sequential update strategy, confidence-aware plane-constrained foot-end update) to enhance stability and mapping accuracy in challenging terrain.
43. Thixotropic properties and rheological optimization of supersulfated cement-based grouting materials for the backfill layer in submarine tunnel behind TBM linings
Core Problem: The rheological properties and thixotropy of supersulfated cement (SSC) grout, crucial for its diffusion and stability in TBM tunnel backfilling, are poorly understood, limiting its application despite its superior resistance to seawater erosion.
Key Innovation: Systematically investigates the effects of cement clinker and calcium carbide slag content on SSC grout's thixotropy and rheological parameters, establishes a constitutive equation for prediction, and demonstrates its superior infiltration and strength (28.7% higher compressive strength) compared to OPC in field tests for submarine tunnel backfilling.
44. A physics-driven hybrid transformer model for hydrologic simulation under nonstationary environmental conditions
Core Problem: Conventional hydrological models with static parameters struggle to accurately represent the nonstationary behavior of hydrological processes driven by climate change and human activities.
Key Innovation: A Physics-Driven Hybrid Transformer Hydrologic Model (PD-HTHM) that integrates an LSTM–Transformer hybrid encoder with a conceptual hydrological model to dynamically generate time-varying parameters, significantly improving robustness and predictive accuracy under nonstationary conditions across various catchments.
45. 3D FE cyclic modelling of monopiles in sand using SANISAND-MS: Calibration and validation from soil element to pile-interaction scale
Core Problem: Accurately modeling the complex cyclic behavior of laterally loaded monopiles in sand for offshore wind applications requires robust 3D FE simulations with well-calibrated constitutive models like SANISAND-MS, but faces challenges in calibration and extrapolation to full-scale conditions, especially in saturated sands.
Key Innovation: Critically evaluates and advances the application of 3D FE modeling for laterally loaded monopiles using the SANISAND-MS model, providing strategies for reliable calibration and highlighting practical considerations, thereby improving the fidelity of advanced FE simulations for offshore monopile design, particularly in dry conditions.
46. Hybrid DEM-FDM modeling of heavy-haul railway transition zone slope effects on ballast particle movement and dynamic track responses
Core Problem: Ballasted tracks in bridge transition zones suffer from weaker stability, intensified structural damage, and accelerated cumulative settlement due to abrupt stiffness changes, necessitating a deeper understanding of macro- and micro-mechanical mechanisms like ballast particle movement and foundation settlement.
Key Innovation: Development and validation of a hybrid DEM-FDM coupled model for ballasted track systems in bridge transition zones to investigate the effects of different slope patterns, demonstrating that reducing the transition zone slope smooths dynamic responses and decreases sleeper/ballast particle displacement amplitudes, providing guidance for performance monitoring.
47. Dissolution and Flow Channeling in Hydrate‐Bearing Sediments: Implications for Permeability
Core Problem: A paradox exists where laboratory-derived effective permeabilities of natural hydrate-bearing sediments are much higher than field observations or theoretical models, challenging efficient methane production and CO2 injection.
Key Innovation: Demonstrates through experiments and a 1D reaction-transport model that methane-free water flow dissolves hydrates and creates flow channels, leading to artificially high measured permeabilities in laboratory settings. It proposes that in-situ permeabilities are much lower, aligning with field data and theoretical models.
48. Diagnosing the Flocculation–Transport Dynamics of Suspended Particulate Matter Using a Two‐Class Population Balance Model and Bayesian Calibration
Core Problem: Accurately simulating and understanding the flocculation-transport dynamics of suspended particulate matter (SPM) in aquatic systems is crucial for water quality and sediment transport, but challenging due to complex aggregation and breakage processes.
Key Innovation: A mechanistic and diagnostic framework combining a two-class population balance equation (TCPBE) model with Bayesian calibration to simulate flocculation-transport behavior, enabling estimation of uncertain model parameters and mechanistic interpretation of flocculation dynamics.
49. Centrifuge modeling-based p-y curves for laterally loaded monopiles in sand considering the effect of pile-soil relative stiffness
Core Problem: Existing p-y models for laterally loaded monopiles in sand often neglect the effect of pile-soil relative stiffness and may not accurately capture load eccentricity effects, leading to less accurate predictions of lateral response.
Key Innovation: Proposes a new p-y model derived from centrifuge tests, featuring a novel formulation for the initial p-y curve slope that accounts for pile-soil relative stiffness, a correction coefficient from direct p-y measurements, and an updated ultimate soil resistance calculation incorporating pile-soil skin friction, demonstrating improved accuracy across various pile dimensions and sand densities.
50. ESA Arctic+ Salinity Product v4: Enhanced Retrievals Near the Ice-Edge
Core Problem: Previous versions of the SMOS sea surface salinity product in the Arctic suffered from contamination and reduced coverage near the ice edge, leading to inaccuracies.
Key Innovation: The ESA Arctic+ Salinity Product v4.0, which incorporates improved algorithms at Level 1 and Level 2 to reduce contamination and correct biases near the ice edge, resulting in a 25% increase in coverage and a 24-43% reduction in RMS difference with in situ measurements compared to the previous version.
51. Detecting supramolecular organic nanoparticles during heat wave
Core Problem: Atmospheric new particle formation is typically hindered by high temperatures, but heat waves are becoming more frequent, raising questions about their impact on particle formation and related public health/climate effects.
Key Innovation: Observed frequent new particle formation events during heat waves, revealing a kinetic pathway for supramolecular nanoparticle production by organic acids, with implications for public health and climate in a warming world.
52. Decoding active sites in high-entropy catalysts via attention-enhanced model
Core Problem: Identifying active sites in high-entropy materials is challenging due to their multiple random sites, hindering the optimization of catalysts.
Key Innovation: Developed an attention-enhanced, multiobjective predictive model to precisely identify active sites and their corresponding overpotentials in high-entropy catalysts, leading to the screening and validation of a high-performance oxygen evolution reaction (OER) catalyst (TiFeNiZn-CoOOH).
53. Comparative analysis of Pn full-waveform inversion and travel-time tomography in imaging upper mantle azimuthal anisotropy
Core Problem: Conventional Pn-wave imaging techniques, based on ray theory, neglect finite-frequency effects, limiting their resolution and accuracy in resolving upper mantle azimuthal anisotropy structures.
Key Innovation: Demonstrated through synthetic experiments that full-wave Pn inversion, which incorporates full-wave finite-frequency sensitivity kernels, significantly improves the resolution and robustness in recovering both velocity and anisotropic parameters compared to conventional travel-time tomography, particularly for sharp gradients and deeper features in tectonically complex settings.
54. Semantic rule-guided three-dimensional reservoir modeling method using an improved multiple-point geostatistics simulation
Core Problem: Traditional Multiple-Point Geostatistics (MPS) methods for 3D reservoir modeling suffer from reliance on manual parameter adjustment, difficulty in uncertainty quantification, and a general lack of geological semantic rule utilization, limiting model reliability.
Key Innovation: Proposed a 3D reservoir modeling method that integrates geological semantic rules (sequence stratigraphy, fault kinematics) into the structural modeling workflow and employs an improved MPS algorithm with automatic domain kernel function selection and rapid dual-dimensional parameter optimization, enhancing geological consistency, coherence, and accuracy in characterizing lithofacies and property distribution.
55. Integrating OBN seismic data and machine learning for enhanced fluid discrimination in pre-salt carbonate reservoirs
Core Problem: Accurately identifying reservoir fluids in ultra-deep pre-salt reservoirs remains challenging due to complex lithology, strong heterogeneity, and limited well data, leading to high fluid identification uncertainties.
Key Innovation: Developed an integrated workflow combining OBN seismic data, pre-stack inversion, AVO analysis, and an innovative LSTM rock physics model (replacing conventional shear-wave modeling) to autonomously learn nonlinear relationships for fluid discrimination, demonstrating superior performance in resolving ambiguous elastic responses and reducing uncertainties in complex pre-salt reservoirs.
56. Urbanization, climate variability, and groundwater dynamics in Eastern India: a mixed-effects modelling approach
Core Problem: Understanding the complex interactions between land cover change, climate variability, and seasonal recharge processes on groundwater levels in rapidly urbanizing regions.
Key Innovation: Used a linear mixed-effects modeling framework to assess the effects of built-up expansion, population growth, and precipitation on seasonal groundwater fluctuations, showing urbanization modifies natural recharge effectiveness.
57. Experimental Study of Transfer Properties in Salt Rocks: Comparison of Diffusion and Permeability Transfers
Core Problem: Limited experimental investigation and comparison of liquid permeability, gas permeability, and diffusion coefficient in salt rocks under hydrostatic stress, especially concerning creep strain effects, which are critical for long-term stability of underground repositories.
Key Innovation: A novel series of laboratory tests characterizing liquid/gas permeability and diffusion coefficient in salt rocks under varying confining pressures, injection fluid pressures, and creep strains, revealing that permeability diminishes irreversibly with increased confining pressure and creep deformation, and that permeation is the predominant transfer mechanism.
58. Investigating Blast-Induced Damage: A Comprehensive Study of Crack Development Profiles Under Decoupled Charges
Core Problem: The challenge of predicting and controlling the extent of blast-induced cracks in rock to optimize blast design and prevent excessive structural damage in mining and tunneling.
Key Innovation: A comprehensive study combining ultra-high-speed photography, 2D Digital Image Correlation, and 3D numerical modeling to quantify blast-induced crack development, propagation velocities, and damage levels under decoupled charges, demonstrating how increasing the decoupling ratio reduces damage.
59. Water transmission resilience analytics informed by hydraulics using connectivity, path diversity, and stability
Core Problem: Existing methodologies for assessing water distribution network resilience often focus on single aspects, lacking an integrated approach to evaluate reservoir-to-tank resilience.
Key Innovation: Developed a resilience assessment framework integrating hydraulic connectivity (graph theory + hydraulics), supply path diversity, and supply path stability into a composite resilience score, providing a holistic evaluation of network performance.
60. Data-driven Time-variant Reliability Analysis using Deep Gaussian Processes
Core Problem: Conventional Time-variant Reliability Analysis (TRA) struggles when statistical properties of stochastic inputs must be inferred from limited, non-stationary measured data, leading to significant unquantified epistemic uncertainty in reliability predictions.
Key Innovation: A comprehensive data-driven TRA framework that integrates multivariate kernel density estimation (KDE) with bootstrapping to non-parametrically model correlated random variables and their epistemic uncertainty, and employs deep Gaussian processes (DGP) as a probabilistic model for complex, non-stationary random processes, systematically quantifying and propagating epistemic uncertainty to yield the full probability distribution of reliability.
61. Conformal Machine Learning for Reliable Anomaly Detection in Industrial Cyber-Physical Systems
Core Problem: Machine learning and deep learning methods for anomaly detection in industrial cyber-physical systems suffer from high false alarm rates (FARs), hindering their trustworthiness and adoption.
Key Innovation: Proposes a novel framework integrating ML-based anomaly detectors with conformal prediction (CP) and a temporal quantile adjustment method to ensure statistical guarantees on predefined FARs, improve detection capability, and enhance interpretability.
62. From Physics to Machine Learning and Back: Part II - Learning and Observational Bias in Prognostics and Health Management (PHM)
Core Problem: Real-world Prognostics and Health Management (PHM) faces challenges like noisy/incomplete sensor data, limited labels, and complex degradation behaviors, limiting the reliability and physical consistency of data-driven models.
Key Innovation: Reviews how physics-informed machine learning addresses PHM limitations by embedding physical knowledge through learning biases and observational biases, enabling a transition from passive prediction to active decision-making and scaling solutions fleet-wide.
63. Investigating the contributions of herbaceous vegetation biomass to soil accretion in Louisiana's coastal deltaic wetlands using airborne imaging spectroscopy
Core Problem: Understanding the relationship between vegetation productivity and soil accretion is critical for assessing the resilience of deltaic wetlands to relative sea-level rise, and landscape-scale quantification of vegetation contributions was needed.
Key Innovation: This study integrates airborne imaging spectroscopy with field measurements to quantify herbaceous vegetation contributions to vertical accretion in coastal wetlands, developing imaging spectroscopy-based products for live above/belowground carbon and aboveground necromass, demonstrating their value for assessing wetland resilience and informing management strategies.
64. SP-KAN: Sparse-sine perception Kolmogorov–Arnold networks for infrared small target detection
Core Problem: Existing infrared small target detection (IRSTD) methods struggle to discriminate dim targets heavily entangled with complex interference due to their fixed activation representations.
Key Innovation: Proposes SP-KAN, a novel Sparse-sine Perception Kolmogorov–Arnold Network, which reformulates IRSTD as a global context modulation problem using sparse nonlinear modules and a sparse-sine perception layer, achieving superior accuracy, robustness, and generalization.
65. Prediction of cutting tool wear in long and complex strata based on field inspection geometry: Insights from a case study of twin earth pressure balance shield tunnelling
Core Problem: Accurate prediction of cutting tool wear in long-distance Earth Pressure Balance Shield (EPBS) tunnelling through heterogeneous, highly abrasive strata is challenging, impacting maintenance planning and operational efficiency.
Key Innovation: Provides the first comprehensive field-based validation of the Japanese Tunnelling Society (JTS) wear prediction method for multiple cutter types across different installation radii in long-distance twin tunnels, demonstrating its reasonable accuracy for maintenance planning (MAE = 4.46 mm) and revealing critical wear mechanisms.
66. Causal mechanisms and parameter optimization strategies of carbon emissions in tunnel boring machine excavation
Core Problem: Minimizing carbon emissions in Tunnel Boring Machine (TBM) excavation is challenging due to uncertainties in operational parameters and complex causal relationships with geological conditions.
Key Innovation: Proposes a hybrid intelligent optimization framework using GraN-DAG to construct a causal network, GPSR to derive parameter relationships, and an Improved Snake Optimization (ISO) algorithm to dynamically adjust TBM operational parameters for minimized carbon emissions under varying geological conditions.
67. Mechanism and design strategy via surfactant hydrophobic tail engineering for enhanced dust control in underground spaces
Core Problem: Designing effective dust suppressants for underground spaces is challenging due to the complex interplay between surfactant structure (hydrophobic tail) and its adsorption/wetting properties on coal dust.
Key Innovation: Reveals a novel remote regulation mechanism where the hydrophobic C-F chain acts as an electron-withdrawing micro-regulator, influencing the distant hydrophilic head and suppressing hydrogen bonding, leading to a bifunctional synergistic design strategy for next-generation dust suppressants with tunable anchoring-wetting balance.
68. A causality-informed and Copula-based framework for predictive assessment of downstream water-quality risks under extreme scenarios
Core Problem: Quantifying the probability of future hypoxic events and understanding the transmission pathways of multi-source stressors affecting downstream dissolved oxygen (DO) dynamics under extreme hydroclimatic conditions.
Key Innovation: A causality-informed and Copula-based framework, including a CA-LSTM water quality prediction model, to enhance interpretability and predictive performance of DO dynamics, quantify tail risks under extreme conditions, and provide early warning for hypoxic events.
69. High-fidelity digital modeling and comparative morphological analysis of Chang’E-5 and Chang’E-6 lunar grains
Core Problem: A lack of direct quantitative links between micro-scale grain morphology, macro-scale regolith structure, and surface mechanical behavior for lunar regolith, which is critical for landing-site evaluation and infrastructure design.
Key Innovation: Reconstructs 3D lunar regolith structures using XRM and a CNN-based segmentation framework, providing quantitative morphological and porosity analyses that directly link micro-scale grain properties to macro-scale regolith structure and mechanical behavior, aiding lunar landing and infrastructure design.
70. Optimization of long-hole raise blasting considering drilling deviations: Theoretical and numerical approaches
Core Problem: Drilling deviations in long-hole raise blasting (LHRB) are often not accounted for, adversely affecting excavation efficiency and progress, particularly for different blasting modes.
Key Innovation: Investigated the effect of drilling deviation on LHRB, theoretically and numerically comparing burn-cut and spherical cartridge blasting modes, and proposed optimization designs combining both modes to balance efficiency and reliability, successfully demonstrating its feasibility in a field test.
71. Rock fracturing using soundless cracking agents produced from freshwater sludge and carbide sludge
Core Problem: Conventional rock fracturing techniques (blasting) have adverse environmental and safety impacts, and existing soundless cracking agents (SCAs) lack sustainability due to reliance on non-renewable resources.
Key Innovation: Developed sustainable soundless cracking agents (SCAs) from freshwater sludge, carbide sludge, and gypsum, demonstrating their ability to achieve high expansive pressure and successfully fracture granite in the laboratory, offering a promising non-explosive alternative for rock fracturing.