TerraMosaic Daily Digest: Mar 28, 2026
Daily Summary
This March 28, 2026 digest distills 18 selected papers from 546 analyzed records. Across coseismic landslides, coral soils, fractured tunnels, and long-creeping rock slopes, the strongest studies argue that hazard emerges from evolving internal structure rather than from forcing alone. The leading papers resolve how segmentation logic improves post-earthquake landslide detection, how gradation and fabric govern liquefaction resistance in reef-derived soils, and how deformation in jointed slopes and disturbed tunnels accumulates through spatially organized damage rather than uniform weakening.
A second theme is methodological. The most convincing papers convert observation into diagnosis: distributed sensing and microseismicity reveal where tunnel damage is concentrating, groundwater recovery is inferred from regional uplift, dual-porosity behavior is linked to reservoir-slope hydraulics, and isotopic partitioning distinguishes meltwater contributions in runoff generation. Taken together, these studies point to a more mechanism-first hazard science, in which monitoring is useful when it identifies the state variable that controls failure propagation.
Key Trends
The strongest papers explain instability by resolving the material state, damage architecture, and support interaction that allow forcing to become failure.
- Instability is being resolved as a spatially organized process: the leading landslide and tunnel studies track how cracks, fractures, and support interaction migrate through the system before macroscopic failure, replacing uniform-strength assumptions with evolving damage fields.
- Geomaterial specificity is becoming central to prediction: coral soils, dual-porosity reservoir soils, and rooted soil-rock mixtures are treated as distinctive media whose fabric and hydraulic response directly shape liquefaction, deformation, and erosion behavior.
- Hazard sensing is moving from event detection to mechanism inference: the most useful monitoring strategies today use segmentation models, distributed sensing, uplift observations, microseismicity, and isotope tracers to recover the process variable that governs hazard evolution, not merely to record that change has occurred.
Selected Papers
This digest features 18 selected papers from 546 papers analyzed.
1. LS-FPSAM: a SAM-based frequency-prompt model for coseismic landslide detection
Core Problem: Post-earthquake response still lacks detection models that can identify landslides rapidly and accurately enough for emergency operations across complex scenes.
Key Innovation: This paper introduces a SAM-based frequency-prompt framework tailored to coseismic landslide detection, pushing segmentation-style remote sensing toward faster emergency mapping of earthquake-triggered slope failures.
2. Factors Influencing the Liquefaction Resistance of Saturated Gravelly Wide-Graded Coral Soils
Core Problem: Liquefaction resistance of coral soils at real engineering sites remains poorly constrained because most prior evidence is based on cleaner coral sand rather than wide-graded coral deposits.
Key Innovation: Large-scale undrained triaxial testing across reef-derived coral soils shows how gradation and site-specific structure shape liquefaction resistance, extending liquefaction assessment beyond idealized coral sand.
3. Temporal and spatial evolution characteristics of axial chain rockbursts in a high-altitude granite TBM tunnel under tectonic effect
Core Problem: Rockburst sequences in tectonically stressed deep tunnels are difficult to diagnose because their temporal migration and precursor microseismic signatures remain poorly organized.
Key Innovation: Using microseismic monitoring in a Himalayan TBM tunnel, the study resolves a three-stage evolution of axial chain rockbursts and identifies shifting event clusters, b-value changes, and correlation-length patterns preceding failure.
4. Investigation of long-term deformation of high rock slope using a combined discrete-continuum modeling strategy: A case study of the left abutment slope at Jinping I Hydropower Station
Core Problem: Long-term deformation of very high jointed rock slopes remains hard to model because joint creep, anisotropy, and scale effects are not captured consistently in a single framework.
Key Innovation: A discrete-continuum creep strategy links joint-scale behavior to slope-scale deformation, reproducing monitored displacement at the Jinping abutment and clarifying fragmentation-controlled long-term instability.
5. Study on the relationship between surface deformation and internal fracturing of tunnel surrounding rock under uniaxial loading with multiple blasting disturbances
Core Problem: Surface deformation is often monitored in tunnels, but the direct relation between visible strain localization and internal crack growth under repeated blasting remains ambiguous.
Key Innovation: Combined DIC and acoustic-emission measurements show strong temporal and spatial coupling between internal fracture activity, surface strain, and energy dissipation, yielding practical instability markers for disturbed tunnel rock.
6. Face instability mechanisms of shield tunnel undercrossing an existing tunnel: Insights from centrifuge model tests and FDM-DEM simulations
Core Problem: Shield face instability beneath existing tunnels is experimentally underconstrained, especially when the pre-existing tunnel is segmented rather than an idealized rigid inclusion.
Key Innovation: Centrifuge tests and coupled FDM-DEM simulations reveal chimney-like failure bending near the existing tunnel and quantify the associated stress redistribution and segment bending moments during undercrossing instability.
7. Physics-data driven approach for multi-objective intelligent prediction and optimization of deep soft rock tunnel stability
Core Problem: Deep water-rich soft-rock tunnels require long-term stability prediction and support optimization, yet conventional analyses are too slow to guide multi-objective decisions in practice.
Key Innovation: By combining hydro-mechanical creep simulation, a surrogate model, and constrained optimization, this framework translates tunnel stability indicators directly into support-design choices under safety-cost tradeoffs.
8. Rapid Land Surface Uplift and Groundwater Recovery Observed During the Syrian War
Core Problem: Groundwater recovery after abrupt changes in water use is rarely observed at regional scale, particularly where field data are nearly absent.
Key Innovation: Remote sensing and conflict-event data reveal that war-driven irrigation collapse in Syria acted as a large-scale reverse pumping experiment, producing measurable uplift, renewed spring flow, and groundwater recovery.
9. A bimodal soil–water retention curve model for dual-porosity soils in the Three Gorges Reservoir hydro-fluctuation belt
Core Problem: Wetting-drying cycles in reservoir hydro-fluctuation belts reorganize pore structure, but current retention models do not capture emerging dual-porosity behavior well enough.
Key Innovation: The paper develops a physically interpretable bimodal SWRC formulation and validates it against landslide-belt soils from the Three Gorges Reservoir, linking pore evolution under cycling to hydraulic response.
10. Finite-strain semi-analytical solution for bolt-rock interaction in large-deformation tunnels
Core Problem: Support design in squeezing tunnels still relies too heavily on small-strain assumptions that misrepresent relative displacement between bolts and surrounding rock.
Key Innovation: A finite-strain sub-ring stress-release method captures bolt-rock interaction more accurately than small-strain approaches, improving prediction of deformation and support demand in large-deformation tunnels.
11. Physical and mechanical properties of root-reinforced soil–rock mixture investigated by CT scanning and direct shear tests
Core Problem: Vegetation-based slope stabilization is widely practiced, yet the microstructural mechanism by which roots reinforce soil-rock mixtures remains poorly quantified.
Key Innovation: CT imaging and direct shear tests show how roots reorganize pore networks and increase rock-soil contact, clarifying the coupled root-rock-soil strengthening mechanism in mixed slopes.
12. Multi-source isotopic tracing reveals meltwater runoff processes and environmental responses in the Yarlung Tsangpo River Basin
Core Problem: The seasonal partitioning of glacier melt, snowmelt, precipitation, and groundwater in major Tibetan rivers remains insufficiently quantified under ongoing climate change.
Key Innovation: High-frequency isotopic tracing and end-member mixing analysis resolve source contributions through the melt season, showing how runoff drivers shift across freeze-thaw stages in the Yarlung Tsangpo Basin.
13. High-efficiency analysis of large-scale train–track–bridge systems based on a localized nonlinear stiffness decomposition method
Core Problem: Large train-track-bridge interaction models are computationally expensive, limiting their practical use in dynamic safety assessment under complex loading.
Key Innovation: A localized nonlinear stiffness decomposition method confines nonlinearity to the wheel-rail subsystem, achieving near order-of-magnitude speedups while preserving system-level dynamic fidelity.
14. Effects of residual foam and defoamer on flocculation efficiency during in situ muck recycling for EPBS tunneling
Core Problem: In situ recycling of EPBS muck is hindered when residual foam and surfactants disrupt flocculation and dewatering, but the mechanism has remained poorly resolved.
Key Innovation: Tank-scale experiments show how residual foam networks and surfactant adsorption suppress flocculation, clarifying how defoamer-flocculant design can stabilize in situ tunneling-muck recycling.
15. CLIMATHUNDERR: experimental database of buoyancy-driven downbursts
Core Problem: Downburst hazard research is constrained by the lack of controlled experimental datasets that resolve the buoyancy-driven structure of damaging winds.
Key Innovation: CLIMATHUNDERR provides a dedicated experimental database for buoyancy-driven downbursts, creating a reusable benchmark for wind-hazard analysis, model validation, and structural loading studies.
16. Measurements of water droplet size distributions in a turbulent wind tunnel
Core Problem: Turbulent droplet behavior at cloud-clear interfaces remains difficult to analyze quantitatively because detailed laboratory distribution datasets are scarce.
Key Innovation: This dataset documents droplet size distributions across turbulent shear and residence times in a controlled wind-tunnel cloud experiment, supporting future hazard-oriented storm-process studies.
17. Integrated assessment of site quality for coastal Casuarina equisetifolia shelterbelts using ground-based modeling and remote sensing
Core Problem: Coastal shelterbelts are important for shoreline protection, but site-quality assessment is often too sparse to guide large-scale, site-specific management.
Key Innovation: Ground measurements, PCA-based soil assessment, and Landsat inversion are combined to map shelterbelt site quality continuously, improving spatial planning for coastal protection forestry.
18. Global Primary and Secondary Microseism Multi‐Decade Geographic Variation, Secular Intensification, and Period Lengthening
Core Problem: Long-term changes in global microseism intensity remain undercharacterized, obscuring how evolving storm-wave climate reshapes the seismic background field.
Key Innovation: By analyzing decades of continuous seismic data across 73 stations, the study documents broad microseism intensification and period lengthening consistent with longer-period, storm-driven ocean-wave forcing.