TerraMosaic Daily Digest: August 28, 2026
Daily Summary
Direct geohazard papers place failure thresholds inside coupled hydraulic, structural, and material states rather than nominal triggers alone. Centrifuge tests on contractive sand slopes show undrained failure did not globalize until the water table intersected the toe, while gassy submarine slopes become drawdown-sensitive through gas exsolution, density, and bedding anisotropy. Regional rockslide mapping in monoclinal strata, probabilistic analysis of unsaturated slopes, reconstruction of legacy-mine collapse, loess-tunnel invert-heave studies, fracture-flow uncertainty analysis, crystalline fault-zone connectivity, sandy-cobble seepage tests, seismic assessment of tunnels and reinforced retaining walls, and the Yebatan hydropower case all treat instability as an emergent property of evolving subsurface architecture, excavation, and flow.
Coastal, climatic, tectonic, and volcanic studies resolve hazard as coupled field behaviour across atmosphere, surface water, groundwater, ecology, and crust. Improved tropical-cyclone representation and a deterministic-stochastic treatment of tide-surge interaction sharpen extreme sea-level estimation, while post-surge aquifer modelling, estuarine salinity metamodeling, saltmarsh observations, dune-recovery theory, plant-structure controls on wave attenuation, bridge-pier scour assessment, and Yellow River water-sediment optimization connect inundation risk to morphodynamics and salinity. Seasonal snow decline, compound heat-precipitation uncertainty, Arctic permafrost-meander regimes, Amazon drought anomalies, vegetation-mediated drought propagation, land-degradation mapping, European heat-stress decomposition, and wildfire ignition attribution show that hydroclimatic hazard pathways are being resolved through structured environmental controls; tectonic and volcanic papers add comparable structural dependence in Hikurangi slow-slip geometry, Alpine anisotropy, earthquake-driven groundwater isotope shifts, and Pacaya gas partitioning.
A substantial secondary stream expands transferable observational and modelling capability without itself establishing geohazard validation. Recurrent methodological moves include preprocessing-aware foundation-model transfer, multispectral compression, explicit spatial grounding in image-text retrieval, physics-guided evapotranspiration and FAPAR products, optical inversion for suspended particulate matter, quadrilateral PFEM and thermo-mechanical MPM for large deformation, copula or surrogate formulations for sediment and stability prediction, and low-power sensing of groundwater flux and soil shear. Together these papers enlarge the analytical toolkit for Earth-system monitoring, but their relevance to geohazards remains conditional on explicit domain application, calibration, and testing.
Key Trends
Five trajectories organize the August 28, 2026 literature: subsurface-state slope thresholds, interaction-aware coastal hazard systems, regime-based climatic and tectonic forcing, dependency-centred infrastructure risk, and calibration-conscious observational methods.
- Failure thresholds are being recast as subsurface state variables: Across sand slopes, submarine sands, excavation-disturbed rockslides, unsaturated slopes, loess tunnels, and seepage-prone rock masses, instability is tied to water-table position, gas phase, hydraulic conductivity, fracture topology, and excavation-induced stress redistribution. The trajectory is toward mechanistic threshold accounting that can support probabilistic mapping and monitoring design.
- Coastal hazard studies are coupling flooding, salinity, morphology, and ecology: The coastal cohort links cyclone structure, tide-surge interaction, aquifer salinization, estuarine salinity fields, saltmarsh deposition, dune recovery, plant-scale wave attenuation, and bridge-pier scour. Risk assessment is therefore moving beyond peak water levels to interaction-aware coastal system behaviour.
- Climate and tectonic forcing are being organized into regime structure: Compound heat-precipitation uncertainty, thermodynamically controlled snow loss, Arctic permafrost-meander equilibria, vegetation-mediated drought propagation, Hikurangi interface geometry, Alpine anisotropy, and earthquake-driven groundwater isotope shifts all map where forcing reorganizes system response. The emphasis is on structured spatial regimes and pathway dependence, not only on monotonic trends.
- Infrastructure risk is becoming explicitly interdependent and operational: Studies of Yebatan hydropower, seismic tunnels, reinforced soil retaining walls, urban power-water networks, shelter water access, and NaTech decision support treat exposure as a coupled service problem rather than an isolated component problem. Countermeasure design is increasingly framed around functional dependence, staging, and recovery constraints.
- Transferable Earth-observation methods favor calibration and physical guidance: Across foundation-model transfer, image-text retrieval, optical sediment inversion, evapotranspiration and FAPAR products, groundwater sensing, and fiber-optic deformation monitoring, the methodological move is to encode realism in preprocessing, physics, geometry, or sensor coupling. These papers strengthen analytic capability, but they should remain distinguished from hazard-tested geohazard evidence unless application is demonstrated.
Selected Papers
The selected papers span direct studies of landslides, coastal flooding and salinization, volcanic degassing, drought, wildfire, seismic structure, and mountain-infrastructure instability, alongside a secondary methods cohort in remote sensing, sensing hardware, and scientific machine learning. The first group provides hazard-tested findings; the second provides transferable capability whose geohazard relevance depends on explicit application and validation.
1. Gas Measurements Reveal Fractional Degassing and Water Interaction During a Flank Eruption of Pacaya Volcano (Guatemala)
Core Problem: The degassing pathway and role of water interaction during Pacaya's 2021 flank eruption were unclear.
Key Innovation: Combines satellite SO₂ flux, UV camera data, and MultiGAS measurements to separate fractional degassing, scrubbing, and groundwater interaction.
2. Global Improvement of Tropical Cyclone Representation Enhances Hindcasts of Extreme Sea Levels Over 68 Years
Core Problem: ERA5 underrepresents tropical cyclones, causing major underestimation of TC-driven storm surges in global hindcasts.
Key Innovation: Injects improved tropical-cyclone wind and pressure structure into ERA5 to produce a 68-year hindcast with much better extreme-sea-level performance.
3. Investigating Triggers of Undrained Slope Instability Using Centrifuge Modeling
Core Problem: It remains unclear which triggers can initiate undrained instability in contractive sand slopes before drained failure conditions are reached.
Key Innovation: Centrifuge tests show global failure occurred only once the water table intersected the toe, indicating undrained responses followed marginal drained instability rather than independently triggering it.
4. Instability of gassy sand slopes during tidal drawdown: effects of fabric anisotropy and gas exsolution
Core Problem: Gas exsolution, anisotropy, and rapid tidal drawdown can trigger offshore sand-slope instability, but their coupled mechanics are poorly represented.
Key Innovation: Builds and validates an anisotropic critical-state model for gassy sands that reproduces unloading collapse and identifies density, gas saturation, bedding, and drawdown as failure controls.
5. Regional hazard assessment of excavation-disturbed translational rockslides in monoclinal strata: a probabilistic framework integrating failure extent and runout
Core Problem: Excavation-disturbed translational rockslides in monoclinal strata need regional hazard mapping that remains reliable under limited subsurface data.
Key Innovation: Integrates field data, stratigraphic prediction, Monte Carlo limit equilibrium, and InSAR validation to map failure probability, extent, and runout at slope-unit scale.
6. A hybrid deterministic-stochastic method for extreme sea level prediction incorporating tide-surge interaction
Core Problem: Reliable extreme sea-level prediction is hindered by nonlinear tide-surge interaction and the limitations of standard extremes methods.
Key Innovation: Develops a crossing-rate framework that treats tide deterministically and surge stochastically, capturing interaction, serial dependence, seasonality, and non-Gaussian behavior.
7. A deep learning framework for probabilistic stability analysis of slopes considering soil spatial variability and unsaturated steady seepage
Core Problem: Probabilistic stability analysis of unsaturated slopes in spatially variable soils under steady seepage is computationally demanding.
Key Innovation: Couples a PI-KAN seepage solver with a two-stage FNO-CNN surrogate and Monte Carlo simulation for fast probabilistic analysis of unsaturated slopes.
8. Dynamic Responses of Groundwater Flow and Salt Transport to Storm Surge Inundation in Unconfined Coastal Aquifers
Core Problem: Post-surge groundwater-flow recovery and salinity evolution in unconfined coastal aquifers are poorly characterized.
Key Innovation: Identifies rapid flow recovery, transient saltwater-wedge retreat, and an upper saline-plume overshoot after surge inundation.
9. Event-Level Attribution of Wildfire Ignition Cause: A Machine-Learning Approach
Core Problem: Wildfire ignition cause is sparsely reported globally, limiting process understanding and hazard management.
Key Innovation: Trains a large event-level ensemble classifier on 1.03 million records to infer whether ignitions were human-caused or lightning-caused.
10. Simulation of nonstationary seabed wave velocity structure incorporating stratum variability and its application to liquefaction assessment
Core Problem: Model seabed wave-velocity structure under stratigraphic variability so liquefaction susceptibility can be assessed more realistically.
Key Innovation: Combines nonstationary seabed velocity-structure simulation with stratigraphic variability for liquefaction-oriented site evaluation.
11. Multi-objective optimal allocation of water and sediment in the Yellow River Estuary: Balancing flood discharge capacity, landform equilibrium and ecological functions
Core Problem: Yellow River Estuary management must balance flood conveyance, delta morphology, and wetland water needs under competing water-sediment demands.
Key Innovation: Creates a multi-objective allocation model that quantifies threshold water and sediment requirements for discharge capacity, landform equilibrium, and ecology.
12. Escalating Uncertainty Under Increasing Risk in Compound Heat and Precipitation Extremes
Core Problem: The evolution of uncertainty in compound heat and precipitation extremes is poorly understood despite rising risk.
Key Innovation: Separates model, scenario, and internal-variability contributions to CHPE uncertainty and identifies global uncertainty hotspots.
13. Emergent Regimes of River Meandering and Permafrost Extent in Arctic Floodplains
Core Problem: Controls on permafrost extent in meandering Arctic floodplains and its sensitivity to warming are poorly constrained.
Key Innovation: Defines a compact regime space showing when river migration and permafrost growth balance, and where warming most strongly accelerates erosion.
14. 3D Subduction Interface Model for the Hikurangi Margin, New Zealand: Implications for Slow Slip Events
Core Problem: Along-strike differences in Hikurangi slow-slip behavior need better structural explanation.
Key Innovation: Builds a new 3D interface model linking megathrust curvature, subducted seamount chains, and spatial patterns of slow-slip patches.
15. Re-Evaluating Inherited SIF Anomaly Detection Methods for the 2024 Amazon Drought: A 3D Convolutional Autoencoder Approach
Core Problem: Different anomaly-detection methods can yield very different views of drought-related vegetation stress in the Amazon.
Key Innovation: Shows that a 3D convolutional autoencoder captures broader and more persistent SIF anomalies than a pixelwise z-score baseline.
16. Long-Term Dynamics of Land Degradation Risk in Arid Northwest China Revealed by an Integrated Risk Index
Core Problem: Long-term spatial dynamics and drivers of land-degradation risk in arid northwest China need integrated monitoring.
Key Innovation: Builds a 1 km integrated land-degradation risk index from multi-source remote sensing and uses it to map trends and drivers from 2001 to 2024.
17. Collapses of legacy mines in a spatial and evolutionary context
Core Problem: How abandoned slate mines evolve toward flexural toppling, rockfall, and sinkhole collapse through coupled structural and environmental controls.
Key Innovation: Combines field inspection, geological analysis, mapping, and dendrogeomorphology to reconstruct collapse evolution in legacy underground mines.
18. A knowledge-based decision support framework for multi-hazard and cascading NaTech risk assessment: Systematic review and multi-criteria evaluation
Core Problem: How to structure decision support for multi-hazard and cascading NaTech risk assessment across fragmented methods and criteria.
Key Innovation: Builds a knowledge-based framework from systematic review and multi-criteria evaluation to organize cascading NaTech risk assessment.
19. Vegetation change reshapes meteorological-hydrological drought characteristics and propagation process in the Yellow River Basin, China
Core Problem: Vegetation change can alter how meteorological drought propagates into hydrological drought, but the basin-scale effects are hard to isolate.
Key Innovation: Frames vegetation change as a driver of drought characteristics and propagation in the Yellow River Basin.
20. Isotopic responses of hot and cold groundwaters to the 2023 Kahramanmaraş earthquakes: Evidence from oxygen-18 (δ¹⁸O), deuterium (δ²H), and tritium (³H, TU)
Core Problem: How the 2023 Kahramanmaras earthquakes altered isotopic signatures in hot and cold groundwaters.
Key Innovation: Uses oxygen-18, deuterium, and tritium evidence to trace groundwater-system responses to major earthquakes.
21. Major engineering geological challenges and countermeasures for the Yebatan Hydropower Station in the upper Jinsha River
Core Problem: Identify the dominant engineering geological challenges affecting construction of the Yebatan Hydropower Station in the upper Jinsha River and how to mitigate them.
Key Innovation: Packages a large, high-relief hydropower case into a challenge-and-countermeasure framework likely useful for hazard management in complex mountain infrastructure corridors.
22. Spatial and Temporal Variation in Drivers of Saltmarsh Eco-Geomorphic Evolution: Field Insights From the St. Lawrence Estuary, Canada
Core Problem: Numerical marsh-evolution models lack geographically diverse field constraints on the drivers of edge erosion and progradation.
Key Innovation: Builds a multi-campaign hydro-eco-geomorphic data set linking tides, sediment supply, deposition, and vegetation traits to contrasting marsh-edge behavior.
23. Thermodynamic Control of Global Seasonal Snow Decline
Core Problem: The relative roles of precipitation change versus warming-driven phase shifts in snowfall decline are poorly quantified.
Key Innovation: Introduces a physically based attribution framework showing thermodynamic phase transition dominates global seasonal snow loss.
24. Linking branch hierarchy to wave attenuation: from perspective of vortex dynamics
Core Problem: The role of branch hierarchy in controlling wave attenuation needs mechanistic explanation.
Key Innovation: Links vegetation branch structure to wave damping through a vortex-dynamics perspective.
25. HySalt: A metamodel for estimating water level and spatial-temporal distribution of salinity in estuarine environments
Core Problem: Fast estimation of estuarine water level and spatiotemporal salinity patterns is needed for operational analysis.
Key Innovation: Presents HySalt, a metamodel aimed at rapid estimation of both water level and salinity distributions.
26. Knowledge discovery for interpretable and variability-aware permeability prediction under sparse sampling in tropical weathered soils: An Amazon case study
Core Problem: Sparse sampling makes permeability estimation in tropical weathered soils difficult, especially when uncertainty and local variability matter.
Key Innovation: Uses conditional-random-field spatial transfer and interpretable machine learning to extend sparse permeability tests into variability-aware predictions for Amazonian weathered soils.
27. Disparities in the identification of vulnerable communities due to design storm methods: A case study in the Vermilion River Watershed, Southern Louisiana, USA
Core Problem: Different design-storm methods may identify different vulnerable communities, complicating equitable flood-risk planning.
Key Innovation: Uses a Louisiana watershed case to compare how storm-design choices reshape vulnerability identification.
28. Quantifying the dependent failure of urban power-water supply systems under earthquake hazards from the perspective of energy transfer
Core Problem: How to represent directional, cascading earthquake failures between urban power systems, water networks, and terminal users beyond binary dependency models.
Key Innovation: Integrates power-flow and hydraulic models in a directional dual-network framework that traces energy transfer through quasi-Monte Carlo seismic damage scenarios.
29. Stability analysis of rock tunnels and caverns under seismic loading using Q-system and discontinuous modelling
Core Problem: Rock tunnels and caverns under earthquake loading need stability evaluation that links rock-mass classification with discontinuous modeling.
Key Innovation: Combines the Q-system with discontinuous numerical modeling for seismic stability assessment of underground excavations.
30. Extension of the particle finite element method towards quadrilateral-dominated meshes for large-deformation geotechnical problems
Core Problem: Particle finite element methods for large-deformation geotechnics can be limited by mesh quality and efficiency.
Key Innovation: Extends PFEM toward quadrilateral-dominated meshes for large-deformation geotechnical analysis.
31. Geometric-topologically constrained modeling of stochastic fracture networks and uncertainty quantification of dominant factors controlling flow behavior
Core Problem: How to generate stochastic fracture networks that honor geometry and topology while quantifying which factors dominate flow uncertainty.
Key Innovation: Adds geometric-topological constraints and uncertainty quantification of dominant controls on fracture-network flow behavior.
32. Progressive failure mechanism of invert heave in soft-plastic loess tunnels under coupled excavation-seepage effects
Core Problem: Explain how excavation unloading and pore-pressure buildup drive invert heave and progressive failure in soft-plastic loess tunnels.
Key Innovation: Integrates theory, physical modeling, and 3D simulation to trace staged deformation growth and identify weak zones under coupled excavation-seepage loading.
33. Porous and Permeable Crystalline Basement: Evidence for Large-Scale Hydraulic Connectivity at Kivetty, Finland
Core Problem: Crystalline basement fault zones are often assumed to be tight, yet some sustain kilometer-scale fluid flow.
Key Innovation: Combines borehole tests, petrophysics, mineral mapping, and XCT to demonstrate highly connected pore networks in altered brecciated fault cores.
34. Imaging the Alpine Collision With Seismic Radial Anisotropy
Core Problem: Crust-mantle deformation styles and slab geometry across the Alps-Apennines-Dinarides remain incompletely resolved.
Key Innovation: Builds a high-resolution 3D shear-wave and radial-anisotropy model from more than 3,300 seismic stations.
35. A Stochastic Approach for Recovery Times of Coastal Dune Vegetation
Core Problem: Post-disturbance dune vegetation recovery times are difficult to quantify because of coupled biotic and abiotic variability.
Key Innovation: Derives an analytical stochastic recovery-time formulation linking vegetation resilience to morphology, forcing, and species traits.
36. Trends in Heat Stress During Heat Waves Across Europe (1979-2023)
Core Problem: Regional drivers of rising European heat-stress severity during heatwaves are not spatially uniform.
Key Innovation: Decomposes observed and reanalysis heat-stress trends into temperature and humidity contributions across Europe.
37. A Copula-Based Regression Method to Predict Sediment Concentration and Load in Rivers Using Streamflow Data
Core Problem: Traditional watershed models for sediment prediction are data-intensive and costly to deploy.
Key Innovation: Proposes a copula-based regression approach that predicts sediment concentration and load from streamflow alone.
38. A digital twin-based dual-limit-state safety assessment for damaged bridge pile-piers under coupled wave-structure-scour interactions
Core Problem: How to track time-varying ultimate and serviceability safety of damaged bridge pile-piers under coupled wave loading and scour.
Key Innovation: Combines a five-dimensional digital twin, multi-source sensing, high-fidelity updating, and dual-limit-state assessment, validated in scaled wave-flume experiments.
39. Derivation of predictable zones and accuracy of deterministic forecast of short-crested waves from sea surface measurements of annular-sector zones
Core Problem: Annular-sector radar measurements are awkward for conventional deterministic short-crested-wave forecasting methods.
Key Innovation: Derives predictable zones for three processing strategies and shows least-squares reconstruction gives the best forecast accuracy.
40. Preprocessing Mismatch and Input Normalisation in Transferring a Multispectral Foundation Model to Marine Surface Segmentation
Core Problem: Foundation models for multispectral imagery can fail badly when preprocessing and normalization differ between pretraining and deployment data.
Key Innovation: Shows that normalization alignment dominates transfer performance and that a simple zero-parameter spectral adaptation can match or beat heavier adapters.
41. A Spectral-Spatial Decoupling and Fusion Network with Multi-Scale Perception for Multispectral Image Compression
Core Problem: Multispectral compression still struggles to model spatial and inter-band redundancy efficiently enough for high-fidelity transmission and storage.
Key Innovation: Introduces separate spectral and spatial processing paths with multi-scale fusion inside an end-to-end learned compression framework.
42. Quantifying interfacial behavior of embedded fiber optic cables for soil shear deformation sensing
Core Problem: Distributed fiber-optic sensing of soil shear is limited by poorly constrained cable-soil interaction under deformation.
Key Innovation: Combines shear tests and theory to define cable-soil interaction states and quantify how anchorage improves strain transfer for OFDR monitoring.
43. Seismic Performance of Geogrid Reinforced Soil Retaining Walls with Tire Facing: A Numerical Study
Core Problem: Waste-tire-faced reinforced soil walls need quantitative guidance on how fill type and geogrid wrapping affect seismic deformation and load transfer.
Key Innovation: Calibrates a 3D model against shaking-table data and compares alternative fillings and wrap configurations under different PGA levels.
44. Geomorphic recovery of the disturbed braided river system in the Western Carpathians, Slovakia
Core Problem: How a braided river system recovers morphologically and ecologically after engineering disturbance and river-training interventions.
Key Innovation: Combines remote sensing, field data, GIS landform mapping, and hydraulic modeling to trace short-term channel recovery and vegetation recolonization.
45. Bridging institutions and communities through Living Labs: Lessons learned from disaster risk reduction in Hulu Langat, Malaysia
Core Problem: How living-lab approaches can bridge institutions and communities in disaster risk reduction practice.
Key Innovation: Distills implementation lessons from a community-based DRR case in Hulu Langat rather than advancing hazard-analysis methods.
46. Dynamic evaluation of evacuation shelter location impact on water access for a seismic event
Core Problem: How evacuation-shelter locations affect emergency water access after a seismic event.
Key Innovation: Dynamically evaluates shelter siting through the lens of post-earthquake water accessibility.
47. An algorithm for the retrieval of particle backscattering, size, and suspended particulate matter in turbid coastal and estuarine waters (MOSAIC)
Core Problem: How to retrieve backscattering, particle size, and suspended particulate matter in turbid coastal and estuarine waters.
Key Innovation: Introduces the MOSAIC retrieval algorithm for optically complex coastal waters.
48. On-demand global Landsat evapotranspiration product: development, evaluation, and dissemination
Core Problem: How to deliver globally accessible on-demand Landsat evapotranspiration estimates at scale.
Key Innovation: Develops, evaluates, and disseminates a global Landsat evapotranspiration product.
49. Explicit spatial localization and task-adaptive balancing for remote sensing image-text retrieval
Core Problem: How to better align spatial content and semantic balance in remote-sensing image-text retrieval.
Key Innovation: Adds explicit spatial localization and task-adaptive balancing to retrieval between remote-sensing images and text.
50. Physics-Guided Machine learning approach for estimating spatiotemporally seamless FAPAR from Landsat surface reflectance
Core Problem: How to estimate seamless FAPAR from Landsat surface reflectance across space and time.
Key Innovation: Uses physics-guided machine learning to generate spatiotemporally seamless FAPAR from Landsat data.
51. Coupled increments of shear and normal stresses trigger slip instability of granite fractures: An experimental investigation
Core Problem: Slip instability on granite fractures depends on coupled shear and normal stress increments that are not well constrained experimentally.
Key Innovation: Uses experiments to show how simultaneous shear and normal stress changes trigger fracture slip instability.
52. The iFLUX sensor for real-time measurement of horizontal groundwater flux magnitude and direction
Core Problem: Long-term in situ measurement of low groundwater flux magnitude and direction is technically difficult and labor intensive.
Key Innovation: Introduces a borehole-deployed microfluidic sensor that measures horizontal groundwater flux magnitude and direction in real time with low power demand.
53. A thermo-mechanical material point method framework for simulating frost-heave-induced large deformation around low-temperature pipelines
Core Problem: How to simulate severe frost heave, pipeline uplift, and ground-surface deformation around low-temperature pipelines under open-system freezing.
Key Innovation: Couples phase-change heat transfer with three-dimensional frost-heave strain in an MPM framework validated against freezing benchmarks and a laboratory pipeline test.
54. Permeability characteristic of sandy cobble soil based on the large-scale vertical seepage test
Core Problem: Quantify how gradation controls permeability, seepage regime, and failure evolution in sandy cobble soils prone to water inrush and collapse.
Key Innovation: Large-scale vertical seepage tests separate continuously graded and gap-graded behavior and introduce an empirical FRI index for Darcy versus Forchheimer flow states.