TerraMosaic Daily Digest: May 2, 2026
Daily Summary
May 2's papers sharpen the spatial unit of hazard evidence. A global flood-vulnerability model resolves exposure at individual buildings by combining terrain, hydrography, structural height, basement depth, and explainable attribution. A landslide study links land-cover trajectories to failure-prone neighbourhoods in mountainous urban terrain, while InSAR hydrogeodesy shows that groundwater recovery can reverse subsidence, partition pressure diffusion across fissured ground, and create infrastructure hazards. Dam safety is treated with similar specificity: ERT inversions are reconstructed by deep networks and checked against self-potential, borehole, piezometric, and laboratory evidence.
The cold-region and underground papers add a material-control axis. Avalanche runout is driven by climate-adjusted snow accumulation; ice-filled joints, frozen saline soils, and frost-susceptible loess are tested or modelled through phase change, suction, bonding, and shear degradation; tunnel and cavern studies quantify isolation layers, smoke extraction, shield attitude, grouted reinforcement, rockbolt load transfer, fracture networks, and hydrogen-lining damage. Remote-sensing and GeoAI contributions are strongest when they recover physically interpretable variables: flood extent under emergency-computing limits, high-frequency glacier velocity, ecosystem-specific burn severity, shoreline migration, subsurface soil-temperature profiles, and geometry-constrained segmentation or referring.
Key Trends
The strongest methodological gains come from matching the model output to the physical unit that carries risk.
- Exposure is resolved at the unit of loss: the flood paper works at buildings; the landslide paper at land-cover transition neighbourhoods; shoreline, contour-rip, and mining-subsidence studies at mapped terrain elements that can be checked or managed.
- Deformation products are paired with forcing: InSAR uplift is interpreted through aquifer recovery and fissure-controlled diffusion; tunnel settlement through excavation geometry; frozen-ground wavefields through phase change; soil-temperature profiles through coupled microwave and thermal infrared cycles.
- Cold-region hazards are controlled by weak interfaces and phase transitions: avalanche runout, ice-filled rock joints, frost heave, floating-ice impacts, frozen saline soil, and unsaturated loess all depend on thermal history, water migration, bonding, and shear degradation.
- Underground engineering papers privilege cross-validated mechanisms: dam defects, tunnel isolation, smoke extraction, shield control, gradient grouting, rockbolt calibration, fracture-network generation, and cavern-lining damage are constrained by paired tests, monitoring, geophysics, or numerical-mechanical models.
- GeoAI contributes when it preserves geometry or physical meaning: emergency flood segmentation, SAM2/LoRA multimodal segmentation, Think2Seg-RS, Geo-R1, physically constrained evapotranspiration, and lithological mapping are valuable because their outputs remain tied to objects, boundaries, energy-water balance, or terrain structure.
Selected Papers
This issue contains 69 selected papers from 1,233 papers analyzed. The strongest cluster links global building-level flood vulnerability, land-cover controls on landslide hazards, InSAR-constrained groundwater-recovery uplift, earth-filled dam defect reconstruction, climate-sensitive avalanche modelling, emergency flood segmentation, and cold-region rock-joint mechanics. The remaining papers extend this evidence base into UAV subsidence monitoring, flood sedimentology, tunnel settlement, wildfire spread and burn severity, shoreline change, frozen-ground mechanics, shield control, fracture-network modelling, cavern-lining integrity, and geotechnical sensing.
1. Global Flood Vulnerability Model: Building-Level Assessment Using Multi-Source Remote Sensing
Core Problem: Flood-risk screening is still often local, categorical, or dependent on field surveys, which limits rapid comparison of building-level exposure across countries.
Key Innovation: The Global Flood Vulnerability Model fuses terrain, hydrography, building height, basement depth, multi-resolution DEMs, LiDAR-trained ConvNeXt height filling, Google Earth Engine delivery, and SHAP attribution to produce physically interpretable building-level vulnerability without local calibration.
2. Land use/land cover controls and neighborhood signatures of landslide hazards in a mountainous urban region of China
Core Problem: Mountainous cities are changing fast, but susceptibility models rarely connect landslide occurrence to neighborhood-scale land-use transitions before and after failure.
Key Innovation: The Lin'an study integrates 2017-2024 LULC trajectories, landslide inventories, machine-learning susceptibility modelling, SHAP interpretation, and transition-path analysis to identify a Disturbance-Degradation-Reconstruction sequence around landslide-prone slopes.
3. Long-Term Surface Uplift Driven by Groundwater Recovery in Xi’an, China: InSAR Constraints on Aquifer Storage and Hydraulic Diffusivity
Core Problem: Urban groundwater recovery can reverse subsidence into uplift, but the hydraulic controls and infrastructure hazards of this process remain much less understood than extraction-driven settlement.
Key Innovation: Sentinel-1 time-series InSAR and well records quantify uplift in Xi'an, estimate aquifer storage and diffusivity, and show how ground fissures partition pressure diffusion while rapid rebound creates seepage-related hazards.
4. ERT-guided reconstruction of defect patterns in earth-filled dams with multi-source validation
Core Problem: ERT can detect anomalous dam zones, but conventional inversions often blur defect geometry under complex seepage conditions.
Key Innovation: A U-shaped deep-learning reconstruction framework learns from Wenner and Schlumberger ERT inversions, tests CNN, KAN, Transformer, and Mamba feature extractors, and validates field dam-defect patterns with soil resistivity, self-potential, borehole, and piezometric evidence.
5. Long term dynamic avalanche modelling under climate change scenarios: A case study in the Italian Alps
Core Problem: Avalanche maps depend on short-period snow accumulation assumptions, yet climate-change effects on accumulation extremes and dynamic runout remain poorly quantified.
Key Innovation: Poli-Aval-2D is driven by climate-scenario snow-accumulation changes in the Italian Alps, linking future H72 shifts to long-term physically based avalanche dynamics and hazard-zone evolution.
6. Gated Lightweight CNN-Transformer Fusion for Real-Time Flood Segmentation on Satellite Internet Terminals Under Triple-Disruption Emergency Conditions
Core Problem: Flood segmentation models are often too computationally heavy for emergency terminals operating with damaged roads, power cuts, and unstable links.
Key Innovation: A gated lightweight CNN-Transformer architecture fuses local boundary detail and global SAR context for Sen1Floods11 flood segmentation on satellite-internet terminals under triple-disruption deployment constraints.
7. Shear mechanical properties of ice-filled rock joints under low-temperature and freeze-thaw cycles
Core Problem: Freeze-thaw cycles degrade ice-filled joints, but direct shear controls on cold-region slope instability remain incompletely quantified.
Key Innovation: Low-temperature direct-shear tests vary freezing temperature, ice thickness, joint roughness, normal stress, and freeze-thaw cycling to resolve shear-strength degradation mechanisms in ice-filled rock discontinuities.
8. A Hierarchical Multi-Scale Denoising Framework for UAV-Derived Digital Subsidence Models in Coal Mining Areas
Core Problem: UAV photogrammetry can map mining subsidence, but multi-scale composite noise obscures subsidence boundaries and weakens parameter extraction.
Key Innovation: A hierarchical denoising framework combines subsidence-value stratification, adaptive DBSCAN outlier removal, and curvature-aware small-scale noise suppression, reducing DSuM error against ground monitoring points.
9. Depositional characteristics of the great flood along the Juma River, North China Plain, in 2023
Core Problem: Flood deposits are critical for reconstructing extreme discharge, but the diagnostic signals of modern flood flow and recession deposits remain site dependent.
Key Innovation: Field sampling after the 2023 Juma River flood links deposit thickness, grain size, pollen, and magnetic susceptibility to flow-stage and recession-stage processes, providing reference criteria for regional paleoflood identification.
10. An InSAR-based monitoring framework for ground settlement management in urban railway tunneling
Core Problem: Ground instrumentation around dense urban tunneling sites is spatially limited and difficult to maintain after construction.
Key Innovation: A PS-InSAR and SBAS-InSAR framework selects stable reference control points, checks consistency with conventional measurements, maps settlement troughs, and identifies deformation-prone zones around a Hong Kong railway tunneling project.
11. The stress-controlled mechanism of deep-buried tunnels based on functional gradient reinforcement in high tectonic stress zones
Core Problem: Annular grouting designs may not exploit the radial stress path of fractured surrounding rock in deep tunnels.
Key Innovation: A functional-gradient reinforcement concept and semi-analytical elastoplastic solution design a radial stiffness-strength gradient, showing how shallow bearing and deeper stress regulation can coordinate tunnel stability.
12. A Modality-Aware Ensemble-of-Experts Model for Wildfire Spread Prediction
Core Problem: Wildfire spread models need to combine heterogeneous remote-sensing predictors while remaining reliable under changing fire behaviour.
Key Innovation: A modality-aware ensemble-of-experts model assigns specialized predictors to different input modalities and fire contexts, improving spread forecasting relative to monolithic deep-learning baselines.
13. Cooperative LoRA with frequency priors for SAM2-Based multimodal remote sensing semantic segmentation
Core Problem: Multimodal segmentation models can fail when SAR, optical, LiDAR, or other sensor streams disagree at boundaries and textures.
Key Innovation: Cooperative LoRA with frequency priors adapts SAM2-based multimodal segmentation to reduce structural cross-modal inconsistency and improve boundary preservation under domain shifts.
14. Bridging semantics and geometry: A decoupled LVLM–SAM framework for reasoning segmentation in optical remote sensing
Core Problem: Large vision-language models can reason over scenes, but their pixel masks are often weakly grounded in remote-sensing geometry.
Key Innovation: Think2Seg-RS decouples linguistic reasoning from mask generation by training an LVLM prompter to drive a frozen SAM model with structured geometric prompts.
15. Geo-R1: Improving few-shot geospatial referring expression understanding with reinforcement fine-tuning
Core Problem: Few-shot geospatial understanding fails when models cannot reason over object-context relations from sparse labels.
Key Innovation: Geo-R1 applies reasoning-centric reinforcement fine-tuning to multimodal geospatial referring, strengthening few-shot localization and interpretation of complex remote-sensing scenes.
16. First mapping of subsurface soil temperature profiles by integrating diurnal cycles of microwave and thermal infrared observations
Core Problem: Subsurface soil temperature controls hydrology, permafrost, frost heave, and ecosystem response, but satellite observations usually capture only surface thermal states.
Key Innovation: Microwave and thermal infrared diurnal cycles are integrated to retrieve subsurface soil-temperature profiles, opening a remote-sensing route for thermal state variables below the land surface.
17. Toward elucidating accessory pigments in intense phytoplankton blooms using hyperspectral satellite remote sensing in support of harmful algal bloom (HAB) monitoring
Core Problem: Chlorophyll products map bloom biomass but often cannot resolve bloom-dominant phytoplankton classes needed for HAB risk interpretation.
Key Innovation: Hyperspectral satellite observations are evaluated for accessory-pigment retrieval, extending HAB monitoring from bulk chlorophyll to bloom composition and potential toxicity indicators.
18. High-Frequency observations of glacier ice velocities at Drang Drung Glacier, Western Himalaya, using a terrestrial time-lapse imaging system
Core Problem: Slow Himalayan glaciers are poorly constrained at seasonal and sub-seasonal timescales because satellite estimates have large uncertainties and sparse cadence.
Key Innovation: A terrestrial time-lapse imaging system provides high-frequency ice-velocity observations at Drang Drung Glacier, resolving seasonal dynamics that support glacier hazard and water-resource assessment.
19. Ecosystem-specific calibration of satellite-derived spectral indices improves the characterization of fire severity
Core Problem: Global spectral-index calibrations can misrepresent fire severity where ecosystem structure controls burn response.
Key Innovation: Sentinel-2 indices are calibrated with field Composite Burn Index plots by ecosystem type, improving fire-severity characterization for post-fire management in Mediterranean landscapes.
20. The Dynamics and Surface Signal of Slab Break‐Off in Continental Settings: Insights From 3D Numerical Modeling
Core Problem: Surface uplift and basin subsidence attributed to slab break-off remain non-unique in geological interpretation.
Key Innovation: A suite of 3D numerical experiments links mantle rheology and lateral tearing velocity to the timing and magnitude of uplift-subsidence signals after continental collision.
21. How Does Heterogeneity Control Strain Localization Patterns in High‐Porosity Rocks?
Core Problem: Shear, dilation, and compaction bands affect rock failure and permeability, but their co-development under microstructural heterogeneity remains difficult to predict.
Key Innovation: An elastoplastic phase-field model separates brittle grain-breakage energy from ductile dissipation and shows how porosity and grain-size heterogeneity organize multiple localization modes.
22. Crustal Responses to the Destruction of Continental Lithosphere: Insights From Radial Anisotropy of the Tanlu Fault Zone, Eastern China
Core Problem: Lithospheric destruction in eastern China reshaped crustal architecture, but its link to present-day seismicity is still debated.
Key Innovation: Joint inversion of receiver functions and surface-wave dispersion images radial anisotropy and lower-crustal mafic intrusions, connecting mantle-derived restructuring to earthquake-promoting asperities.
23. Impacts of urbanization on temperatures and agricultural drought in Africa based on Urban Heat Drought Index from 2000 to 2023
Core Problem: African cities can intensify thermal stress and agricultural drought, but combined urban-heat drought metrics are still emerging.
Key Innovation: An Urban Heat Drought Index derived from 2000-2023 data quantifies urbanization-linked temperature and drought impacts in Greater Cairo and Nairobi.
24. A Methodological Framework for High-Latitude Coastal Classification Using ICESat-2 and Explainable Machine Learning
Core Problem: Sea ice, snow, and seasonal darkness limit optical mapping of high-latitude coasts, where climate sensitivity is high.
Key Innovation: ICESat-2 photon-counting data and explainable machine learning are combined to classify coastal types and diagnose controls on high-latitude coastal morphology.
25. The role of seasonal vegetation dynamics in shaping river delta channel networks and morphodynamics
Core Problem: Delta models often simplify vegetation despite its control on sediment trapping, channel routing, and land-loss resilience.
Key Innovation: Delft3D FM is coupled with a dynamic multi-species vegetation model to simulate seasonal biogeomorphic feedbacks in delta channel-network evolution.
26. Modelling mountain pine beetle-impacted forest wildland fire fuel distributions from remotely piloted aircraft systems imagery and point clouds
Core Problem: Wildfire fuel mapping needs 3D vegetation structure that captures insect disturbance and canopy fuel changes.
Key Innovation: RPAS photogrammetry and LiDAR point clouds are compared for deriving fuel-distribution metrics in beetle-impacted forests in Jasper National Park.
27. NatShore: Automated nation-scale shoreline extraction and change analysis using multidecadal cross-mission remote sensing data
Core Problem: Large-scale shoreline monitoring is limited by tidal variability, cross-mission inconsistencies, and manual processing.
Key Innovation: NatShore provides an automated, open-source workflow using Landsat and Sentinel-2 time series for nation- to continental-scale shoreline extraction and change analysis.
28. High-resolution mapping of saturated soil hydraulic conductivity across China’s drylands
Core Problem: Sparse soil samples and coarse environmental covariates limit hydrologic modelling of infiltration and runoff in drylands.
Key Innovation: A high-resolution mapping framework integrates multi-source covariates and machine learning to estimate saturated hydraulic conductivity across China's drylands.
29. Quantifying contour rip line error in altered landscapes using high resolution digital elevation models
Core Problem: Contour ripping is used to manage overland flow and erosion, but its direction and off-contour error are rarely quantified automatically.
Key Innovation: A high-resolution DEM workflow derives rip-line direction, rip-line error, and furrow slope, enabling surface-water and landform-stability assessment in altered landscapes.
30. Unvegetated corridors nonlinearly regulate hydrological connectivity and sediment delivery in a microtidal marsh
Core Problem: Marsh resilience depends on sediment delivery, but invaded vegetation can reroute flow and starve interior platforms.
Key Innovation: The study shows how unvegetated corridors nonlinearly regulate hydrological connectivity and sediment delivery, clarifying vegetation-management effects on microtidal marsh resilience.
31. Micro-fracturing and deterioration mechanisms of sandstone in the Nankan grotto, China: Critical role of biotite weathering
Core Problem: Weathering-induced micro-fracturing threatens sandstone heritage sites, but the mineralogical driver is often poorly isolated.
Key Innovation: Mineralogical, geochemical, and microstructural analyses identify biotite weathering as a critical mechanism for sandstone deterioration in the Nankan grotto.
32. Detection of frozen soil structures using Rayleigh waves: Insights from thermo-mechanical coupling and MASW analysis
Core Problem: Underground construction using artificial ground freezing needs reliable monitoring of frozen-curtain thickness and continuity.
Key Innovation: Thermo-mechanical coupling simulations and MASW analysis link water-ice phase change to Rayleigh-wave responses, supporting non-invasive frozen-ground detection.
33. Constitutive modelling of frozen saline soil incorporating generalized cohesion and generalized density
Core Problem: Frozen saline soils show nonlinear strength and deformation because ice and salt crystallization jointly control bonding and density.
Key Innovation: Generalized cohesion and generalized density variables are introduced to unify confined-compression and triaxial shear behaviour in frozen saline soil.
34. Risk analysis on the impact of floating ice at the water intake of power plants during cold wave weather
Core Problem: Floating ice during cold waves can threaten water intakes, but drift-path models often neglect wave effects.
Key Innovation: A Lagrangian transport framework couples sea ice, tides, and waves to trace floating-ice source regions and quantify short-term intake-impact risk.
35. A two-system heat transfer approach to soil freezing: Numerical simulation of frost heave considering local thermal non-equilibrium effects
Core Problem: Engineering frost-heave tools must represent phase change, water migration, and frozen-fringe evolution without excessive simplification.
Key Innovation: A two-system heat-transfer framework models saturated silty-soil freezing and frost heave under local thermal non-equilibrium between soil particles and pore water.
36. Investigation on the thermo-hydro-mechanical behavior of unsaturated loess constrained by frost heave displacement under unidirectional freezing actions
Core Problem: Freezing-induced deformation of channel subsoil can compromise lining stability, especially in unsaturated loess.
Key Innovation: Unidirectional freezing tests quantify how temperature boundary, gradient, and initial water content govern thermo-hydro-mechanical response and frost-heave development.
37. The development and application of the intelligent detection anchoring (IDA) system for underground engineering
Core Problem: Deep underground construction needs faster field characterization of rock-mass structure and mechanical parameters.
Key Innovation: The intelligent detection anchoring system uses digital drilling to acquire structural and mechanical information in support-design workflows for underground engineering.
38. Study on seismic performance of pressure-resistant isolation layer based on shaking table tests
Core Problem: Tunnel rubber isolation layers can be compressed by static loads, reducing seismic isolation efficiency.
Key Innovation: Material tests, numerical analysis, and shaking-table experiments evaluate a composite pressure-resistant isolation layer with separate compression-resistance and energy-dissipation components.
39. Experimental investigation on smoke control and smoke extraction efficiency of two-point extraction system under longitudinal sloped tunnel fires
Core Problem: Tunnel slope changes smoke movement and extraction efficiency, but two-point extraction performance is not well constrained.
Key Innovation: Scaled fire tests vary slope and exhaust velocity to quantify smoke-control and extraction-efficiency regimes for longitudinally sloped tunnels.
40. Dynamic risk perception-based evacuation path optimization framework in metro fire scenarios
Core Problem: Static evacuation routing can fail when fire, smoke, visibility, and crowd density evolve during a metro emergency.
Key Innovation: DSCA-Evac couples FDS smoke distributions with dynamic path-cost optimization, updating route choice according to evolving fire-risk perception.
41. Three-dimensional dynamic Green’s function of layered half-space with irregular interfaces excited by harmonic point load
Core Problem: Irregular soil interfaces can substantially modify vibrations, yet many Green's-function solutions assume regular layers.
Key Innovation: A semi-analytical 3D dynamic Green's-function method combines boundary integrals, plane-wave expansion, discrete wavenumbers, and generalized reflection/transmission matrices.
42. Hypoplastic modeling of calcareous sand considering particle breakage under anisotropic consolidation and various loading paths
Core Problem: Calcareous sands can crush under anisotropic consolidation, but many constitutive models assume isotropic normal consolidation.
Key Innovation: A stress-state-dependent breakage law is embedded in a hypoplastic framework to reproduce calcareous-sand response under multiple consolidation ratios and loading paths.
43. Modal damping contributed by infills with horizontal frictional sliding joints in infilled frames under seismic excitations
Core Problem: Infills with sliding joints dissipate seismic energy, but their damping contribution is difficult to represent in structural models.
Key Innovation: A component-based macro element and amplitude-dependent damping formulation estimate modal damping and seismic response for infilled frames.
44. Adaptive wavelength-dependent dispersion curve extraction
Core Problem: Surface-wave site characterization is limited by modal ambiguity and erroneous dispersion picks.
Key Innovation: An adaptive wavelength-dependent extraction method improves the reliability of dispersion data used for geotechnical inversion.
45. Shield machine attitude prediction and control based on outcome-oriented and numerical fusion technologies
Core Problem: Shield attitude control must respond to changing geology and construction states in real time.
Key Innovation: A CatBoost-NSGA-III-IPSO-GRU model fuses monitoring data, numerical simulation, and outcome-oriented parameters to predict and correct shield machine attitude.
46. Projection-based low-dimensional parameterization method for three-dimensional discrete fracture networks through generative adversarial network
Core Problem: DFN inversion is ill posed because 3D fracture geometry is high dimensional and field observations are sparse.
Key Innovation: WGAN-GP models generate 2D projections, dip-angle maps, and projection-distance maps, then reconstruct high-fidelity 3D fracture networks from low-dimensional latent variables.
47. Calibration of a 2D built-in rockbolt model using pre-yield axial load distribution data for fully grouted rebar rockbolts
Core Problem: Common numerical rockbolt models can misrepresent pre-yield axial load development, weakening underground excavation design.
Key Innovation: Fiber-optic laboratory and tunnel-scale data are used to calibrate a built-in elasto-plastic interface rockbolt model through bond shear stiffness.
48. Evaluation of lining damage for hydrogen storage caverns: A cross-scale model
Core Problem: Hydrogen storage caverns need cross-scale evaluation of seal-lining deformation and cracking under cyclic storage expansion.
Key Innovation: A finite-difference and discrete-element cross-scale model captures concrete cracking and seal-lining interaction to evaluate lining damage in lined rock caverns.
49. Automatic establishment of discrete fracture network from outcrop: Estimate of fracture length and density via reinforcement learning
Core Problem: Inferring 3D fracture length and density from 2D outcrops is biased by orientation, truncation, and censoring.
Key Innovation: A reinforcement-learning method based on Monte Carlo Tree Search optimizes DFN parameters from outcrop observations, automating fracture-network establishment.
50. Development and validation of physically constrained machine learning for improving remote sensing-based evapotranspiration estimation
Core Problem: Satellite ET models can be accurate in-sample but physically inconsistent and weak under extrapolation or extremes.
Key Innovation: AutoML compares unconstrained, input-constrained, and loss-constrained models, showing that physical ET constraints improve generalization and water-balance consistency.
51. Tracing changes in subsurface water storage through a novel satellite-based time-series of far-red solar-induced fluorescence quantum efficiency
Core Problem: Drought monitoring needs variables that reflect plant physiological stress and subsurface water storage rather than only surface greenness.
Key Innovation: A satellite time series of far-red SIF quantum efficiency is used to trace changes in subsurface water availability and vegetation water stress.
52. Lithological mapping in semi-arid shallow-covered areas using a multi-scale temporal dual attention network
Core Problem: Vegetation and surface cover mask bedrock signatures in semi-arid regions, weakening lithological maps used for terrain and resource assessment.
Key Innovation: MSTDA-Net uses multi-scale temporal convolution and dual attention to extract implicit lithology-vegetation relationships from multi-source time series.
53. Accurate identification of rare earth mining area from low-resolution remote sensing images: a cloud-based approach integrating super-resolution and semantic segmentation
Core Problem: Large-area mining supervision often relies on low-resolution imagery that blurs mine boundaries.
Key Innovation: A cloud-based super-resolution and semantic-segmentation workflow improves rare-earth mining-area detection for ecological monitoring and resource oversight.
54. Annual and diurnal temperature cycle modelling of a merged multi-annual ECOSTRESS and Landsat land surface temperature dataset
Core Problem: High-resolution LST dynamics are needed for heat, drought, permafrost, and hydrological applications but remain sparsely sampled.
Key Innovation: Merged ECOSTRESS and Landsat observations are fitted with annual-diurnal cycle models to generate spatially detailed LST-cycle parameters.
55. Spectroscopy-based models outperform pedotransfer functions for estimating soil hydraulic properties
Core Problem: Soil hydraulic properties are expensive to measure and pedotransfer functions can be weak in heterogeneous soils.
Key Innovation: Spectroscopy-based models are compared with pedotransfer functions, showing stronger estimation of water-retention and hydraulic variables.
56. Seismic capacity of cut-and-cover frame structures with isolation layer: A parametric study
Core Problem: Underground frame structures require validated seismic-isolation design under parametric variations.
Key Innovation: Centrifuge-test-validated numerical models evaluate EPS isolation-layer effects on seismic capacity across structural and soil parameters.
57. Experimental and parametric study on the flexural behavior of novel corrugated steel plate–mold bag concrete composite joints for tunnel support
Core Problem: Corrugated steel plate support joints can fail in bending because of insufficient stiffness.
Key Innovation: Full-scale tests and numerical simulations quantify the flexural behaviour and failure law of corrugated steel plate-mold bag concrete composite joints.
58. Coupled analysis of contact pressure and deformation in deep-buried, ultra-large-section jacking prestressed concrete cylinder pipe in loess strata: Field test and numerical simulation
Core Problem: Deep-buried large-section pipe jacking has nonlinear pipe-soil-slurry contact pressures that are hard to predict.
Key Innovation: Field testing and numerical simulation track coupled contact pressure and deformation in China's largest-section JPCCP jacking project.
59. A coupled elastoplastic damage subloading surface model for rock-like geomaterials subjected to triaxial cyclic compression
Core Problem: Rock-like materials accumulate elastoplastic damage under cyclic loading, but constitutive models often separate plasticity and damage poorly.
Key Innovation: A coupled elastoplastic damage subloading-surface model represents cyclic compression response and progressive stiffness degradation.
60. DEM modelling of the axial behaviour of underground pipelines buried in unsaturated coarse-grained soils
Core Problem: Axial pipeline response depends on suction-controlled unsaturated soil behaviour that is rarely resolved in DEM models.
Key Innovation: DEM simulations analyse buried-pipeline axial behaviour in unsaturated coarse-grained soils, clarifying particle-scale load transfer.
61. Study on the influence of concrete core filling on the dynamic response and crack resistance of offshore prestressed pipe piles during hammering piling
Core Problem: Offshore prestressed pipe piles can crack during hammer piling, but full-scale dynamic evidence is limited.
Key Innovation: Field pile-driving tests and ultrasonic damage detection compare conventional open-ended and core-filled pipe piles to quantify crack resistance and dynamic response.
62. Field-scale estimation of geomaterials porosity in transportation embankments using elastic wave velocities
Core Problem: Compaction density controls pavement and embankment stability, but field porosity estimation remains difficult.
Key Innovation: Elastic-wave measurements across field compaction zones are used to estimate dry unit weight and porosity in transportation embankments.
63. Moisture-dependent mechanical response and multi-scale failure mechanisms of crushed stone cemented by a rapid-setting subaqueous grout
Core Problem: Reinforcing fractured rock masses and granular subgrades in water-rich environments is hampered by grout dispersion and weak bonding.
Key Innovation: A rapid-setting subaqueous polymer-cement grout is tested across moisture states to reveal multiscale failure mechanisms and bonding performance.
64. Coupled effects of suction and cyclic loading on pullout response of geogrid reinforcement in unsaturated sand
Core Problem: GRS structures are sensitive to matric suction and cyclic loading, but their coupled pullout response remains poorly understood.
Key Innovation: Pullout tests examine suction, compaction moisture, and cyclic-load effects on geogrid-soil interaction in unsaturated sand.
65. Static fatigue effects on hydraulic fracture propagation in deep shale: Insights from true triaxial experiments and acoustic emission characterization
Core Problem: Deep shale fracturing is affected by time-dependent fatigue and stress state, but true-triaxial evidence remains limited.
Key Innovation: True-triaxial experiments and acoustic-emission characterization reveal static-fatigue controls on hydraulic-fracture propagation.
66. Full-scale compaction and density homogeneity of GMZ bentonite–sand blocks as a high-level radioactive waste barrier in China by X-ray computed tomography
Core Problem: Bentonite-sand blocks for radioactive-waste barriers require full-scale density homogeneity verification.
Key Innovation: X-ray computed tomography evaluates compaction and density uniformity in GMZ bentonite-sand blocks used for high-level waste containment.
67. Simultaneous measurement of soil thermal properties using distributed temperature sensing
Core Problem: Soil thermal properties govern heat and moisture transport, but conventional point measurements are sparse.
Key Innovation: Distributed temperature sensing is used to measure soil thermal properties along fiber-optic cables, improving spatial coverage for subsurface thermal characterization.
68. Influence of mineral–water interaction on water nanodynamics in clays
Core Problem: Pore water in clay-rich soils behaves differently from bulk water, affecting swelling, strength, and transport.
Key Innovation: NMR relaxation mapping and nanodynamic modelling quantify mineral-water interaction effects on pore-water behaviour in clays.
69. Revisiting the self-sealing behavior of compacted bentonite in an annular technological void
Core Problem: Repository bentonite barriers must self-seal technological voids, but sealing evolution under constrained geometry remains uncertain.
Key Innovation: The study revisits compacted bentonite self-sealing in an annular void, informing buffer performance for underground waste isolation.