TerraMosaic Daily Digest: August 13, 2026
Daily Summary
Direct geohazard studies extend from static susceptibility mapping toward sequence-aware and exposure-aware risk accounting. A conterminous-United-States framework links landslide susceptibility, temporal exceedance probability, and economic consequence, and finds that spatial risk variability is controlled more by exposure than by hazard alone, with mean risk rising by about 170% from 2-year to 10-year scenarios. In the Türkiye-Syria sequence, probabilistic landslide hazard is conditioned jointly on coseismic disturbance and subsequent atmospheric-river rainfall, while Armenian earthquake scenarios show that losses depend strongly on local site conditions, masonry vulnerability, and urban concentration. Postseismic analysis of the 2024 Wushi earthquake likewise shows that coseismic stress change and months of afterslip promoted later aftershocks, extending hazard beyond the main rupture.
Mechanistic studies across slopes, floods, and tectonic systems resolve instability through structure, fluids, and catchment connectivity. The Kuayanzi rockslide reconstruction ties failure to lithologic contrast, basal weakening, and runoff concentration into rear fissures; 66 years of Piedmont records show a larger summer share of shallow-landslide events despite stable or declining mean summer rainfall, implicating convective intensification and supporting threshold recalibration. Potential landslide dams are shown to generate outburst floods capable of reactivating downstream slopes, and a permeable-asphalt debris-flow brake materially reduces runout in flume tests. Earthquake and volcanic papers add that near-lithostatic pore pressure can facilitate flexural reverse faulting, Condor Seamount seismicity is consistent with an inflating upper-mantle magma reservoir, and fault-controlled caving can overturn empirical expectations of mine stability.
Hydrologic, climatic, and transferable-method papers deepen this picture while remaining careful about what has and has not been validated for hazards. Flood hydraulics studies show that large wood and macro-roughness produce regime-dependent backwater and turbulence, typhoon precipitation products systematically underdetect extreme rain, reservoir impoundment reshapes event-scale flood-sediment responses, and groundwater hysteresis increasingly decouples network extent from discharge as snow shifts to rain; wildfire and climate papers further identify distinct pre-fire soil-moisture and vegetation-water trajectories across Africa, rapid growth of heat-driven moisture demand in the southern Amazon, and mesoscale convective systems as the dominant control on changes in the tropical rainfall seasonal cycle. The transferable cohort then emphasizes calibrated uncertainty, geometry-aware matching, structured radar and photogrammetric reconstruction, physically constrained retrieval, parsimonious or error-aware hybrid modeling, and faster atmospheric computation, but these advances should be read here as enabling methods unless hazard validation is demonstrated within the study itself.
Key Trends
August 13.s papers converge on geohazard analysis that is more cascade-aware, materially resolved, and validation-conscious, while transferable methods emphasize physical plausibility and calibrated uncertainty rather than benchmark inflation.
- Risk frameworks are decomposing susceptibility, triggering, and exposure: Across the United States landslide assessment, the Türkiye-Syria earthquake-rainfall sequence, Armenian seismic scenarios, and regional flood or climate-risk mapping, these studies repeatedly separate where systems are predisposed to fail, what perturbation activates them, and how exposed assets convert hazard into loss. This marks a shift away from single-score hazard products toward modular risk architectures.
- Failure diagnosis is becoming materially and structurally explicit: The Kuayanzi rockslide, Radim sinkhole, flexural reverse faulting, Wushi postseismic sequence, Condor Seamount unrest, and shaft water-inrush studies all show that hidden structure, pore pressure, degradation history, and channelized connectivity can dominate failure timing and style. Empirical or coarse representations are repeatedly shown to miss these controls.
- Hydrometeorological analysis is increasingly regime-dependent: Piedmont landslides, Tokar-gap wave climatology, typhoon precipitation evaluation, snowmelt-runoff onset mapping, groundwater hysteresis under snow-to-rain transition, and tropical rainfall seasonal-cycle change all resolve hazard behavior by season, flow state, forcing regime, or event geometry rather than by long-term means alone. The common outcome is that upper-tail behavior and timing matter more than bulk climatology.
- Cascade propagation is now a primary analytical target: Several direct hazard papers focus not just on initiating events but on how disturbance migrates through linked systems: earthquake-to-rainfall landslide escalation, landslide-dam outburst floods reactivating downstream slopes, reservoir impoundment reorganizing flood-sediment responses, and multi-fault ruptures increasing cumulative structural damage. Hazard analysis is therefore moving toward chain-of-effects accounting rather than isolated event description.
- Transferable methods prioritize calibration, structure, and falsification of apparent skill: The methodological cohort emphasizes calibrated spatial uncertainty, physics-guided wind reconstruction, structure- and geometry-guided image matching, Toeplitz-structured TomoSAR, physically constrained SIF retrieval, parsimonious differentiable hydrology, symbolic-regression checks on error compensation, and source-stratified validation against domain mismatch. The shared trajectory is not larger models for their own sake, but better constrained inference and more defensible evaluation.
Selected Papers
The selected papers divide into direct geohazard studies and transferable sensing, mechanics, hydrologic, climate, and AI methods. Direct contributions span landslides, floods, earthquakes, wildfire, volcanic unrest, coastal hazards, and recovery, whereas the transferable cohort contributes observation, reconstruction, uncertainty, and modeling tools whose geohazard utility is prospective unless demonstrated in the paper itself.
1. From landslide susceptibility to risk assessment in the conterminous United States
Core Problem: Large-scale landslide mapping in the United States needed to move beyond susceptibility toward explicit risk assessment.
Key Innovation: From the title alone, the paper appears to build a national-scale framework that links landslide susceptibility to risk assessment across the conterminous United States.
2. An Inflating Upper Mantle Magma Reservoir Beneath the Tip of the Condor Seamount, Azores Revealed by Rapid-Response Ocean-Bottom-Seismometer Data
Core Problem: What caused the Condor Seamount earthquake crisis and whether it reflects ongoing magmatic inflation.
Key Innovation: Rapid-response ocean-bottom seismometers plus ML phase picking resolve stable upper-mantle seismicity consistent with an inflating magma reservoir.
3. Modeling the combined effects of the 2023 Türkiye-Syria earthquake and an Atmospheric River event on landslide hazard
Core Problem: How the 2023 Turkiye-Syria earthquake and a later atmospheric-river event jointly changed landslide hazard across the affected region.
Key Innovation: Combines satellite observations with physics-based probabilistic modeling of coseismic and hydrologic drivers and extends the framework toward seasonal forecasting.
4. Elevated fluid pressure in compression facilitates flexural reverse faulting
Core Problem: Standard flexural-faulting models underpredict the depth and slip of compressional reverse faulting during plate flexure.
Key Innovation: Shows through numerical modeling that near-lithostatic pore-fluid pressure can enable deep flexure-driven reverse faulting and amplify seismic hazard.
5. Failure mechanism and dynamic characteristics of the Kuayanzi rockslide at the Qutang Gorge, the Three Gorges Reservoir, China
Core Problem: The Kuayanzi rockslide's failure mechanism and runout dynamics needed reconstruction after the 2025 event.
Key Innovation: Integrates field surveys, UAV photogrammetry, lab testing, and RAMMS back-analysis to link lithologic contrast, runoff concentration, and pore-pressure buildup to rapid rockslide failure and impact dynamics.
6. Spatiotemporal distribution, climatic factors, and seasonal precipitation patterns characterizing 66 years of widespread shallow landslide events in Piedmont (northwestern Italy)
Core Problem: Long-term regional links between widespread shallow-landslide occurrence, seasonality, rainfall climatology, and warning-threshold performance were poorly quantified.
Key Innovation: Analyzes 66 years of Piedmont landslide events to reveal changing seasonal coupling, summer convective intensification effects, and where early-warning thresholds need recalibration.
7. Scenario-Based Seismic Risk Assessment of Six Armenian Cities: Integration of Hazard, Exposure, and Vulnerability Models
Core Problem: Armenia lacked a harmonized scenario-based seismic risk framework integrating hazard, exposure, and vulnerability for major cities.
Key Innovation: Builds a city-scale framework combining deterministic fault scenarios, Vs30 site effects, GIS exposure, and locally calibrated vulnerability functions within ELER.
8. Interplay of Coseismic Rupture, Afterslip, and Aftershocks Associated with the 2024 Mw 7.1 Wushi Earthquake
Core Problem: The spatial and temporal interaction of rupture, afterslip, and triggered aftershocks in the 2024 Wushi sequence needed resolution.
Key Innovation: Uses InSAR slip inversion and Coulomb stress analysis to show how coseismic slip and subsequent afterslip promoted later aftershocks on nearby faults.
9. Experimental study on an improved debris-flow dewatering brake using permeable asphalt concrete and a method for predicting its mitigation effectiveness
Core Problem: Debris-flow dewatering brakes needed a more practical design and a predictive method for estimating mitigation performance.
Key Innovation: Tests a permeable-asphalt dewatering brake in flumes and derives an empirical energy-based model linking brake properties to post-installation runout reduction.
10. A multi-fidelity CNN framework fusing numerical simulation and discretized upper bound limit analysis for limited-width slopes with spatially variable strength
Core Problem: Three-dimensional random-field slope reliability analysis is too computationally expensive when using only high-fidelity mechanical simulations.
Key Innovation: Fuses FLAC3D and discretized upper-bound limit analysis in multi-fidelity CNN workflows that improve factor-of-safety and failure-probability prediction at much lower cost.
11. The geomechanical paradox of the Radim sinkhole: A forensic investigation into fault-controlled caving in “Stable” stopes
Core Problem: A fault-controlled sinkhole developed above supposedly stable stopes, creating a geomechanical paradox that required forensic explanation.
Key Innovation: From the title alone, the paper appears to reconstruct how fault structures triggered caving and sinkhole formation despite nominally stable underground openings.
12. Multi-scale reconstruction of snow-avalanche frequency and vegetation structure using dendrogeomorphology, satellite imagery, and UAV photogrammetry
Core Problem: Long-term avalanche frequency and associated vegetation-structure changes were difficult to reconstruct across scales.
Key Innovation: Combines dendrogeomorphology, satellite imagery, and UAV photogrammetry to reconstruct snow-avalanche frequency and vegetation structure.
13. Cascading flood impacts of potential landslide dams revealed by InSAR and numerical simulation
Core Problem: Potential landslide dams and their downstream flood consequences needed better cascading-hazard analysis.
Key Innovation: From the title alone, the paper appears to combine InSAR detection with numerical simulation to reveal flood impacts from potential landslide dams.
14. Flow resistance parameterization for large wood in two-dimensional depth-averaged models: Effects of flow condition, blockage, and sediment
Core Problem: How to parameterize flow resistance from large wood in 2D depth-averaged models across blockage, flow, and sediment conditions.
Key Innovation: Regime-dependent drag and Manning calibration from flume experiments using Gaussian-process optimization, showing sediment-shaped morphology strongly changes effective resistance.
15. Flume experiments on macro-roughness elements: Hydraulic response and potential influences on flood risk and habitat diversity
Core Problem: How macro-roughness restoration elements change backwater rise and flow heterogeneity under different hydraulic configurations.
Key Innovation: Extends a semi-analytical backwater model across multiple structure types and shows width and Froude number dominate hazard-relevant water-level effects.
16. Physically Consistent Reconstruction of Sparse Scatterometer Ocean Surface Wind Fields Based on Physics-Guided Generative Learning
Core Problem: How to reconstruct physically consistent ocean-surface wind fields from sparse scatterometer observations.
Key Innovation: A physics-guided generative network uses physics-consistency scoring and loss terms to recover coherent wind vectors even during typhoons.
17. Tokar gap wind jet climatology and Red Sea wave trends (1979-2024)
Core Problem: How Tokar Gap winds control Red Sea wave climatology, extremes, and long-term hazard trends near the Saudi coast.
Key Innovation: Builds the first 45-year jet-wave climatology and shows upper-tail wind strengthening does not translate directly into coastal wave-height trends.
18. Machine learning-driven flood susceptibility mapping in the coastal frontier of Feni District, Bangladesh
Core Problem: Flood-prone coastal frontier zones in Feni District lacked comparative machine-learning susceptibility mapping.
Key Innovation: Builds and compares SVM, logistic regression, and KNN flood-susceptibility models using 14 environmental predictors, with SVM performing best.
19. Climate-risk hotspot mapping from rainfall extremes and land use dynamics
Core Problem: How rainfall extremes and land-surface change jointly organize regional climate-risk hotspots was insufficiently mapped.
Key Innovation: Combines long-term extreme-rainfall indices with NDVI and land-surface temperature in a self-organizing-map framework to map shifting Odisha climate-risk hotspots.
20. A Photogrammetric Simulation Framework for Rockfall Change Detection with Statistically Validated Measurement Noise
Core Problem: Rockfall change-detection methods lack realistic ground truth and noise-controlled datasets for quantitative evaluation.
Key Innovation: Creates a photogrammetric simulator that removes controlled rock blocks and injects statistically validated measurement noise to benchmark rockfall detection workflows.
21. Trajectories of Microwave-Based Soil and Vegetation Water Content Underlying Wildfire Dynamics in Africa
Core Problem: Large-scale pre-fire and post-fire coupling between soil moisture, vegetation water content, and wildfire dynamics was insufficiently understood.
Key Innovation: Combines long-term microwave soil-moisture and vegetation-optical-depth data with fire observations to reveal region-specific pre-fire trajectories and post-fire ecohydrological coupling shifts across Africa.
22. Performance Evaluation of IMERG and GSMaP Hourly Precipitation Products for Landfalling Typhoon Rainfall in China
Core Problem: The reliability of IMERG and GSMaP hourly precipitation estimates during landfalling typhoons in China was insufficiently characterized.
Key Innovation: Builds a distance-season-diurnal evaluation framework showing systematic underestimation of heavy typhoon rain and identifying where bias correction is needed before hazard use.
23. Mechanisms of water-inrush mitigation under alternative grouting strategies during shaft sinking: Solidification-enabled stress-seepage-damage coupling
Core Problem: Different grouting strategies for shaft sinking in faulted aquifer settings lacked a mechanistic basis for performance comparison.
Key Innovation: Integrates grout diffusion, solidification, and excavation disturbance in a coupled stress-seepage-damage model and clarifies source weakening, channel blocking, and disturbance regulation roles.
24. Seismic response analysis of building portfolios via multi-GPU parallel computing considering structure-soil-structure interaction
Core Problem: The effect of structure-soil-structure interaction on portfolio-scale residential seismic response in dense urban settings was insufficiently quantified.
Key Innovation: Uses multi-GPU 3D nonlinear simulations to show how spacing, frequency content, and site conditions alter drift, acceleration, base shear, and damage patterns.
25. Cumulative damage of SDOFs under the multiple acceleration sequences induced by multi-fault ruptures
Core Problem: The structural damage implications of multi-fault-rupture ground motions with multiple acceleration sequences were not well constrained.
Key Innovation: Compares recorded and synthetic multiple- versus single-sequence motions and shows larger short-period damage demands, with the Park-Ang index most sensitive.
26. Efficient assessment for tropical cyclone wind hazard under future climate based on probability density evolution method integrated with change of probability measure
Core Problem: Future tropical-cyclone wind hazard is difficult to evaluate efficiently under changing climate conditions.
Key Innovation: Integrates the probability density evolution method with change of probability measure to estimate future tropical-cyclone wind hazard more efficiently.
27. Seasonal transitions and model-inferred interactions of gully erosion susceptibility using explainable machine learning and multi-period remote sensing
Core Problem: Seasonal variability and interacting controls on gully-erosion susceptibility were insufficiently captured in susceptibility mapping.
Key Innovation: Uses explainable machine learning with multi-period remote sensing to infer seasonal transitions and predictor interactions in gully-erosion susceptibility.
28. Cascade-reservoir impoundment reshapes event-scale flood-sediment responses upstream of the Three Gorges Reservoir: evidence from rainfall patterns and driver attribution
Core Problem: How cascade-reservoir impoundment reshapes event-scale upstream flood-sediment responses and driver attribution was insufficiently understood.
Key Innovation: From the title alone, the paper appears to attribute changes in flood-sediment response to rainfall-pattern effects and cascade-reservoir regulation.
29. Rapid Increases in Heat-Driven Extreme Moisture Demand in the Southern Amazon
Core Problem: Why extreme vapor pressure deficit is rising so rapidly in the southern Amazon and how far it may intensify.
Key Innovation: Observation-based attribution and scenario projections show extreme heat is amplifying drought stress and fire-favorable conditions faster than mean drying.
30. Observations-Driven Spatial-Attention LSTM for Multi-Depth Soil Moisture Prediction and Dynamic Soil Hydrological Process Connectivity Quantification
Core Problem: How to predict multi-depth soil moisture while also diagnosing time-varying vertical hydrologic connectivity.
Key Innovation: A spatial-attention LSTM improves deeper-layer soil-moisture prediction and converts attention weights into interpretable connectivity diagnostics.
31. A global high-resolution dataset of snowmelt runoff onset timing from Sentinel-1 SAR, 2015-2024
Core Problem: How to map snowmelt runoff onset timing globally at high resolution despite cloud and observational limitations.
Key Innovation: Creates an 80 m global Sentinel-1 SAR runoff-onset dataset for 2015-2024 validated against ground sensors.
32. Observed fingerprint of global warming exposes model biases in regional climate attribution
Core Problem: Regional attribution is hindered by model biases and difficulty separating forced warming from internal variability in observations.
Key Innovation: Derives observation-based forced and internal SAT patterns to expose where CMIP6 ensembles misrepresent regional emergence and variability.
33. Calibrated spatial uncertainty for Earth observation foundation models via Matérn-motivated latent stochastic regularization
Core Problem: Earth observation foundation models lacked spatially coherent and calibrated uncertainty estimates, especially out of distribution.
Key Innovation: Adds a Matérn-motivated latent stochastic regularization layer to frozen EO embeddings, substantially improving spatial uncertainty calibration under holdout and deployment settings.
34. High-efficiency reconstruction and discontinuity extraction of rock mass outcrops via Planar-based 3D Gaussian Splatting and Triangle-Wise Clustering
Core Problem: Fast, high-quality reconstruction of rock outcrops and discontinuity extraction remained challenging for field-scale rock-mass analysis.
Key Innovation: Uses planar-based 3D Gaussian Splatting plus triangle-wise clustering for efficient outcrop reconstruction and discontinuity extraction.
35. A dependency-guided symbolic regression framework for overcoming error compensation in hybrid AI-physical models
Core Problem: Hybrid AI-physical models can appear accurate for the wrong reasons because error compensation masks structural problems.
Key Innovation: Introduces dependency-guided symbolic regression to expose and reduce error compensation in hybrid AI-physical models.
36. Fault-controlled uplift segmentation and crust-mantle heterogeneity in the topographic evolution of the Sındırgı Segment (Western Anatolia) ruptured during the 2025 M6.1 earthquake sequence
Core Problem: How fault segmentation and crust-mantle structure jointly control uplift and landscape evolution along the Sindirgi Segment.
Key Innovation: Combines geomorphic indices, drainage metrics, knickpoints, river-profile inversion, and Moho variation to infer spatially variable uplift history.
37. Dynamic Evolution of Forearc Subsidence Controlled by Slab Geometry
Core Problem: Which subduction parameters control forearc subsidence, curvature, and shortening during free slab sinking.
Key Innovation: A 4D analogue experiment suite showing slab geometry and bending radius dominate forearc deformation across parameter choices.
38. GPU-accelerated finite-element method for the three-dimensional unstructured mesh atmospheric dynamic framework
Core Problem: How to accelerate a 3D unstructured-mesh atmospheric dynamic framework on GPUs.
Key Innovation: Reorganizes data layout and finite-element kernels to deliver order-of-magnitude whole-model speedups on GPUs.
39. An argument for parsimony in differentiable hydrologic models
Core Problem: Whether differentiable hydrologic models need high architectural complexity to perform well or remain interpretable.
Key Innovation: A cross-basin evaluation argues simpler hybrid models match complex ones while avoiding misleading process claims.
40. Spatial Domain Mismatch Between Field Plots and GEDI Inflates Aboveground Biomass Model Accuracy in a Sudanian Savanna Woodland
Core Problem: Fusing field plots with GEDI can create misleading biomass model accuracy when the reference datasets sample different spatial domains.
Key Innovation: Demonstrates that apparent skill largely came from between-source separation and argues for source-stratified validation rather than standard spatial cross-validation alone.
41. Bridging Individual-Tree and Stand-Scale Aboveground Biomass Estimation for Chinese Fir Using LiDAR and Machine Learning
Core Problem: Reliable biomass estimation across scales is difficult because UAV LiDAR and regional ALS capture different structural detail.
Key Innovation: Bridges individual-tree and stand-scale biomass estimation by combining UAV LiDAR, simulated sparse LiDAR, Sentinel-2 features, and TabPFN modeling with uncertainty propagation.
42. Var-ANN Calibration of FY-3C VASS Temperature Profiles: Evaluation over the Tibetan Plateau and Application to WRF Precipitation Simulation
Core Problem: Temperature profiles over the Tibetan Plateau are too biased and sparse for robust NWP support.
Key Innovation: Combines variational correction with an ANN to calibrate FY-3C VASS profiles and shows improved radiosonde agreement plus potential WRF precipitation benefits.
43. Structure-Based Feature Representation for Robust Multi-Modal Image Matching
Core Problem: Multi-modal image matching remains unreliable under strong radiometric and geometric differences between sensors.
Key Innovation: Introduces a structure-focused matching pipeline with local normalization, spatially balanced keypoints, and a new max-index HOG descriptor.
44. Comparison and Analysis of Four Terrestrial Water Storage Monitoring Models: A Case Study of the Loess Plateau
Core Problem: Different TWS monitoring approaches over the Loess Plateau produce conflicting trends because hydrologic signals are entangled with non-loading deformation.
Key Innovation: Compares GLDAS, GRACE, GNSS, and joint inversion products and explicitly diagnoses how loess collapse, compaction, and mining subsidence can contaminate TWS interpretation.
45. Toeplitz-structured deep unfolding network for TomoSAR 3-D reconstruction
Core Problem: High-quality TomoSAR 3D reconstruction needed a more efficient and structured deep-unfolding formulation.
Key Innovation: Introduces a Toeplitz-structured deep unfolding network for TomoSAR 3D reconstruction.
46. Learning from noisy pose priors: Geometry-guided routing attention for robust image matching
Core Problem: Image matching degrades when noisy pose priors mislead geometric attention mechanisms.
Key Innovation: Introduces geometry-guided routing attention that learns from noisy pose priors to make matching more robust.
47. A Cost-Effective Probe for Monitoring Groundwater-Surface Water Interactions
Core Problem: How to measure groundwater-surface water exchange cheaply enough for dense monitoring networks.
Key Innovation: A low-cost multi-depth temperature probe reproduces reference flux estimates with small error at far lower marginal sensor cost.
48. Groundwater hysteresis increasingly decouples flowing network length from streamflow as snow shifts to rain
Core Problem: Why flowing stream-network length becomes less tightly coupled to streamflow as snowmelt shifts toward rainfall.
Key Innovation: Distributed simulations and geochemical evidence identify groundwater hysteresis as the mechanism decoupling network extent from discharge.
49. Changes in the tropical rainfall seasonal cycle dominated by mesoscale convective systems
Core Problem: How mesoscale convective systems will reshape the seasonal cycle of tropical rainfall under climate change.
Key Innovation: High-resolution simulations attribute projected seasonal-cycle intensification and delay mainly to mesoscale convective systems.
50. A Semi-Empirical Method for Estimating All-Sky Photosynthetically Active Radiation from Sentinel-2 for High-Resolution Land Surface Analysis
Core Problem: Existing PAR products are too coarse to align with fine-resolution Sentinel-2 land-surface observations.
Key Innovation: Builds a semi-empirical all-sky PAR estimator from Sentinel-2 aerosol, water-vapor, cloud, and solar-geometry information.
51. Application of Temporal Satellite Imagery to Assess Ecological Resilience: A Case Study in the Qianshan Region of the Northeast Forest Belt
Core Problem: Long-term ecological resistance and recovery to disturbance were hard to quantify consistently from satellite time series.
Key Innovation: Uses BFAST breakpoints and a geographic-detector framework to map resilience and its drivers from MODIS NDVI time series.
52. Data Quality Analysis of Wet Atmospheric Temperature Profiles from the YunYao Meteorological Constellation Radio Occultation
Core Problem: The quality and applicability of YunYao radio-occultation wet-atmosphere temperature profiles were undocumented.
Key Innovation: Provides radiosonde-based error characterization showing strong mid-tropospheric usability but degraded near-surface performance needing stricter QC.
53. Analytical model considering flow path tortuosity for evaluating permeation-induced reinforcement effects of jet grouting piles in shallow-buried sections of a mountain tunnel
Core Problem: Quantitative evaluation of grout permeation effects in shallow-buried mountain-tunnel reinforcement has been lacking.
Key Innovation: Builds a tortuous-flow analytical grout-permeation model and embeds it in simulation to quantify diffusion, permeability, strength gain, and settlement reduction.
54. Hydraulic-history effects on the strength and pore structure of compacted claystone: direct comparison between hydraulically processed and directly compacted states
Core Problem: Engineering design did not adequately account for how drying-wetting history changes claystone pore structure and strength beyond dry density and water content alone.
Key Innovation: Directly compares hydraulically processed and directly compacted states and shows hydraulic history drives path-dependent pore evolution, suction hardening, and wetting degradation.
55. Field investigation on strength and stiffness behaviors of a weathered granite
Core Problem: In-situ strength and stiffness of heterogeneous weathered granite were difficult to quantify reliably for design.
Key Innovation: Combines SBPM, PMT, and SPT testing to derive depth trends, loading-rate effects, and empirical links between pressuremeter parameters and SPT values.
56. Deformation-controlled earth pressure evolution in metro station structures during service-stage groundwater rebound
Core Problem: At-rest earth-pressure theory does not capture deformation-controlled stress redistribution during service-stage groundwater rebound in metro stations.
Key Innovation: Combines tests, simulations, and theory to derive a deformation-dependent effective earth-pressure formulation for laminated pile-wall systems under rebound.
57. Community-led co-design in disaster recovery: Enabling conditions for practice
Core Problem: The enabling conditions that make community-led co-design effective in disaster recovery practice were not clearly synthesized.
Key Innovation: Identifies the conditions that support community-led co-design as a recovery practice, based on the title's focus on implementation enablers.
58. FSIF-PCNN: A physics-constrained neural network for full-spectrum solar-induced chlorophyll fluorescence reconstruction
Core Problem: Full-spectrum solar-induced chlorophyll fluorescence reconstruction required a physically constrained learning framework.
Key Innovation: Proposes a physics-constrained neural network for reconstructing full-spectrum solar-induced chlorophyll fluorescence.
59. A data-driven numerical simulation method for dynamic water grouting in fractured media
Core Problem: Dynamic water grouting in fractured media is difficult to simulate accurately using purely mechanistic methods.
Key Innovation: Proposes a data-driven numerical simulation method for dynamic water grouting in fractured media.
60. Precipitation amount versus precipitation phase: a threshold in precipitation-amount change governs the shift in their relative influence on snowmelt dynamics
Core Problem: The relative influence of precipitation amount versus phase on snowmelt dynamics shifts under climate change, but the governing threshold was unclear.
Key Innovation: Identifies a threshold in precipitation-amount change that controls when precipitation amount overtakes phase in influencing snowmelt dynamics.
61. FKDM: a frequency-guided Koopman dynamics model with heterogeneous mixture-of-experts for robust daily runoff forecasting under non-stationarity
Core Problem: Daily runoff forecasting under non-stationarity remained difficult for standard data-driven models.
Key Innovation: Introduces a frequency-guided Koopman dynamics model with a heterogeneous mixture-of-experts architecture for robust runoff prediction.
62. Stress-induced anisotropy and fabric evolution of unsaturated soil-rock mixtures under true triaxial conditions
Core Problem: The evolution of stiffness and damping under induced anisotropy was not well captured for geomaterials under complex stress paths.
Key Innovation: Uses directional stress-path experiments and an extended empirical model to account for anisotropy effects on dynamic soil properties; the supplied title and abstract appear mismatched on material details.
63. Extended One-Dimensional Elastic Visco-Plastic Model for Clays Considering Over-Consolidation and Thermal Effects
Core Problem: Soft clay constitutive models did not adequately integrate temperature, creep, structuration, and anisotropy effects.
Key Innovation: Develops and validates an extended thermal elastic visco-plastic model for clay behavior under variable temperature paths.