TerraMosaic Daily Digest: Mar 29, 2026
Daily Summary
This March 29, 2026 digest distills 30 selected papers from 1518 analyzed records. The strongest contributions do not treat slope failure as an isolated event; they trace how hazard is transferred through terrain, infrastructure, and time. The leading papers link landslide activity decay to downstream debris-flow supply, recast regional detection through DEM-aware transformer models, and show that engineering instability in tunnels, liquefiable soils, and rockfall-prone systems is best understood as coupled interaction among material state, structure, and evolving demand.
A second theme is operational. Dynamic deformation monitoring, InSAR, explainable machine learning, joint probability models, and multi-hazard optimization are being used not simply to label hazardous locations, but to identify when exposure becomes actionable. Across the set, the most convincing studies are those that convert broad hazard categories into mechanism-specific pathways of amplification, transfer, and consequence.
Key Trends
Today’s strongest papers explain geohazards as transfer systems, in which instability evolves across connected materials, processes, and infrastructures rather than remaining confined to a single slope or trigger.
- Hazard chains are displacing single-event narratives: the best papers follow sediment transfer, retrogressive expansion, secondary hazard generation, and infrastructure interaction through time, showing that consequence often depends more on propagation than on initial triggering.
- Susceptibility and exposure modelling are becoming dynamic: DEM-aware transformers, deformation-informed mapping, future forcing scenarios, and explainable models are shifting hazard assessment away from static inventories toward state-aware and forward-looking prediction.
- Engineering geohazards are increasingly treated as coupled systems: tunnel faces, soft-rock supports, liquefiable soils, rockfall loads, and network-scale multi-hazard metrics are all analyzed through interaction among structure, material behavior, and evolving external demand.
Selected Papers
This digest features 30 selected papers from 1518 papers analyzed.
1. A new DEM-integrated dual-branch swin transformer model for landslide detection along the China Sichuan-Tibet railway
Core Problem: Automated landslide detection along the Sichuan-Tibet Railway remains difficult because steep terrain and complex geological conditions degrade image-only recognition models.
Key Innovation: The study integrates DEM information with a dual-branch Swin Transformer, improving landslide detection in one of the most hazard-exposed railway corridors in the world.
2. Chain effects of landslide activity intensity decay on landslide sediment transfer and debris flow activity
Core Problem: The way post-earthquake landslide activity decays through time, and how that decay propagates into sediment transfer and debris-flow activity, remains poorly resolved.
Key Innovation: Long-term monitoring shows that shifts in landslide activity intensity reorganize sediment supply and debris-flow behavior after the Wenchuan earthquake, clarifying chain effects beyond the initial trigger.
3. Large-strain finite element analyses of a retrogressive landslide triggered by pile driving in sensitive clays: the case of the 1978 Rigaud landslide in Québec
Core Problem: Retrogressive landslides in sensitive clays remain difficult to reproduce mechanistically because triggering and post-peak degradation are strongly path dependent.
Key Innovation: Large-strain Eulerian finite-element analyses reproduce the Rigaud landslide and show how pile driving can initiate localized shear bands that evolve into large-scale retrogression.
4. Quantitative evaluation and wide-area simulation of post-failure debris flows using the depth-integrated particle method
Core Problem: Regional simulation of post-failure debris flows remains difficult when many possible source areas must be treated within a single computational domain.
Key Innovation: The paper validates the depth-integrated particle method from flume to regional scale and demonstrates how a single framework can simulate wide-area post-failure debris-flow propagation.
5. Field investigations on large-scale instability triggered by the Chenghai-Binchuan fault zone, northwestern Yunnan, China
Core Problem: Large paleo-landslides along active fault corridors are still incompletely linked to the kinematics and chronology of the structures that generated them.
Key Innovation: Field mapping and dating show how long-lived fault motion in the Chenghai-Binchuan zone organized repeated large-scale instabilities and landslide-dammed lake development through time.
6. Preliminary inferences on the August 5, 2025, catastrophic debris flow in Dharali, India: insights from the satellite imageries and geomorphological analysis of the Kheer Gad watershed
Core Problem: Catastrophic debris flows in steep Himalayan watersheds remain hard to reconstruct quickly enough to separate source, entrainment, and channel-amplification processes.
Key Innovation: Satellite imagery and geomorphological analysis of the Kheer Gad watershed identify how rainfall, moraine failure, and channel morphology combined to amplify the 2025 Dharali debris flow.
7. Critical state and static liquefaction resistance of sand with low-plastic fines via equivalent void ratio concept under varied stress histories
Core Problem: Static liquefaction resistance of sands with low-plastic fines remains difficult to compare across stress histories because state variables are not normalized consistently.
Key Innovation: The study uses an equivalent void ratio framework to reconcile critical-state behavior and static liquefaction resistance of sand-fines mixtures under varied loading histories.
8. Characteristics of rockburst triggered by weak dynamic disturbance: insights from structural model tests
Core Problem: Rockbursts triggered by weak dynamic disturbance remain difficult to explain because the path from subtle perturbation to violent release is rarely observed experimentally.
Key Innovation: Structural model tests reveal how weak disturbances reorganize fracture development and stress redistribution before rockburst occurrence, improving interpretation of non-obvious triggering paths.
9. Acceleration of the Gradenbach-Eggerberg rock slope deformation
Core Problem: Acceleration phases in large alpine rock-slope deformation remain difficult to diagnose because long records of movement, erosion, and river interaction are rarely synthesized together.
Key Innovation: This study analyzes the accelerating deformation of the Gradenbach-Eggerberg slope, clarifying how monitored motion and river interaction combine to maintain a long-lived rock-slope hazard.
10. Seismic slope stability: a detailed review of recent advances in fundamental science, modelling and mitigation strategies
Core Problem: Seismically induced slope failure remains a persistent problem because mechanistic knowledge, modeling practice, and mitigation strategy are still fragmented across disciplines.
Key Innovation: This review synthesizes recent advances in seismic slope science, bridging triggering physics, numerical modeling, and mitigation design in one consolidated framework.
11. Analysis of triggering factors behind the October 2023 South Lhonak GLOF event in the Sikkim Himalaya using multiple remote sensing data
Core Problem: The South Lhonak GLOF highlighted the need to separate the remote-sensing evidence for lake-condition change, trigger timing, and cascade amplification in high mountain basins.
Key Innovation: Multiple remote sensing datasets are integrated to reconstruct the trigger conditions and geomorphic controls behind the October 2023 South Lhonak outburst flood.
12. Landslide susceptibility assessment under future seismic and precipitation scenarios: a case study of the 2014 Mw 6.2 Ludian earthquake zone, Yunnan, China
Core Problem: Susceptibility studies rarely treat future seismic and precipitation forcing jointly, even though their coupled effect governs many mountain-slope failures.
Key Innovation: This study projects landslide susceptibility in the Ludian earthquake zone under future rainfall and seismic scenarios, extending mapping from present-day correlation to scenario-based forecasting.
13. Assessing landslide susceptibility with dynamic deformation monitoring and explainable machine learning: a case study in Longhua County, China
Core Problem: Static conditioning layers dominate landslide susceptibility work, leaving dynamic deformation signals underused and model interpretation limited.
Key Innovation: Dynamic deformation monitoring is integrated with explainable machine learning to improve operational prioritization and expose the drivers behind predicted susceptibility patterns.
14. Unsupervised deep learning for environmental risk monitoring: Landslide detection from multi-resolution remote sensing imagery
Core Problem: Transferable landslide-detection models remain hard to build because labeled inventories are limited and image resolution varies strongly across remote sensing sources.
Key Innovation: This paper evaluates self-supervised representation learning for landslide segmentation across multi-resolution imagery, showing how unsupervised pretraining can improve environmental risk monitoring.
15. Multi-parameter seismic metrics for detection and classification of rock and ice-rock avalanches
Core Problem: Seismic monitoring can capture avalanches remotely, but precursor and low-amplitude stages are still hard to distinguish from ambient noise.
Key Innovation: A multi-parameter seismic metric improves detection and classification of rock and ice-rock avalanches, including precursor stages that standard seismic methods often miss.
16. Estimation of excavation-induced high-damage zone in super-large span tunnels with microseismic monitoring technology
Core Problem: Excavation damage zones remain difficult to delineate in super-large span tunnels, even though they control the actual working state of the surrounding rock.
Key Innovation: This study uses microseismic monitoring to estimate the excavation-induced high-damage zone, turning event-density information into a practical tunnel-damage diagnostic.
17. Experimental and theoretical investigation of face failure and ground collapse during slurry pressure-balanced shield tunneling in saturated sand
Core Problem: Face failure and associated ground collapse in slurry pressure-balanced shield tunneling through saturated sand remain insufficiently constrained for safe pressure control.
Key Innovation: This paper combines experiments and theory to clarify how face failure develops and how collapse propagates during slurry pressure-balanced shield tunneling in saturated sand.
18. A joint probability model for multi-hazard intensity in earthquake-induced rockfall scenarios
Core Problem: Reliability assessment for earthquake-induced rockfall remains limited by the absence of quantitative models linking seismic excitation to rockfall intensity jointly.
Key Innovation: The study builds a copula-based joint probability framework for earthquake-induced rockfall intensity, enabling multi-hazard load characterization for structural reliability analysis.
19. Integrating engineering and non-engineering mitigation for secondary hazards at Mount Marapi: Sabo dams and irrigation intakes for debris flow control
Core Problem: Post-eruption debris-flow management often separates river engineering from community and irrigation-system considerations, weakening real-world mitigation effectiveness.
Key Innovation: This paper links sabo-dam design, intake adaptation, and non-structural management into a combined mitigation strategy for secondary debris-flow hazards at Mount Marapi.
20. Field characterization of areas in İskenderun affected by liquefaction during the 2023 Kahramanmaraş earthquake
Core Problem: The 2023 Kahramanmaraş earthquake exposed severe liquefaction in İskenderun, but high-quality subsurface field case histories are still needed for calibration and mitigation design.
Key Innovation: Field CPT and SCPT investigations characterize the deposits and deformation patterns in the most affected zones, strengthening the empirical basis for liquefaction engineering in urban coastal settings.
21. Evaluating liquefaction exposure of road networks to support decision-making
Core Problem: Infrastructure managers need network-scale liquefaction exposure measures, yet most assessments remain focused on individual sites rather than system disruption.
Key Innovation: This paper introduces exposure indicators for road networks under hazard scenarios and return-period shaking, connecting liquefaction manifestation to transport-system decision support.
22. Dynamic seismic risk assessment of personnel entrapment at the building level in urban areas
Core Problem: Search-and-rescue operations need building-level estimates of trapped occupants, but dynamic seismic risk models for entrapment remain limited.
Key Innovation: The study develops a dynamic assessment framework for earthquake-induced personnel entrapment, moving post-event rescue planning closer to building-specific risk intelligence.
23. Multi-geohazard susceptibility assessment and influencing factors in Zhejiang Province, China: a machine learning approach
Core Problem: Collapses, landslides, and debris flows often co-occur, yet susceptibility assessment usually treats them separately and leaves cross-hazard drivers undercharacterized.
Key Innovation: Machine learning is used to map multi-geohazard susceptibility at provincial scale and quantify the factor structure governing collapse, landslide, and debris-flow patterns together.
24. Quantifying flash flood inundation and assessing damage using satellite earth observations: the case of 2022 flash flood in Bangladesh
Core Problem: Flash-flood response requires rapid spatial quantification of inundation and damage, but event-scale assessments remain fragmented across data types.
Key Innovation: Satellite earth observations are fused to reconstruct inundation dynamics and associated damage from the 2022 Bangladesh flash flood, yielding a more explicit event footprint for impact analysis.
25. A multihazard assessment framework integrating single and coupled scenario analysis for droughts, floods, and landslides
Core Problem: Hazard assessments still underrepresent interactions between droughts, floods, and landslides, despite their tendency to amplify one another in coupled scenarios.
Key Innovation: This framework evaluates both single and coupled hazard scenarios, expanding multihazard assessment from parallel mapping to interaction-aware analysis.
26. SBAS-InSAR for sustainable development goals (SDGs)—the impact of extreme precipitation on the foothill's stability of the South Taihang Mountains, China
Core Problem: Extreme rainfall can induce widespread foothill deformation, but the surface response outside urban flood zones often remains poorly monitored and weakly interpreted.
Key Innovation: SBAS-InSAR is used to isolate precipitation-driven deformation in the South Taihang foothills, connecting radar-observed motion to rainfall-triggered slope-stability concerns.
27. Evolution of the OpenQuake Engine: Enhanced capabilities, collaborative development, and global adoption
Core Problem: Seismic hazard workflows need open and extensible engines that can keep pace with changing engineering and risk-assessment requirements.
Key Innovation: This paper documents how OpenQuake expanded its modeling and collaborative capabilities, reinforcing its role as operational infrastructure for seismic hazard and risk analysis.
28. An enhanced system reliability framework for regional seismic risk assessment considering inter-structural correlations
Core Problem: Regional seismic risk studies often ignore correlations between structural responses, which can bias system-level loss and reliability estimates.
Key Innovation: This framework explicitly incorporates inter-structural correlations into regional seismic reliability analysis, improving system-level earthquake risk assessment.
29. A novel dynamic disaster risk assessment of Urban Built Environments: an application to flood and earthquake
Core Problem: Urban disaster risk in built environments changes with exposure and occupancy through time, but most risk models remain effectively static.
Key Innovation: The study develops a dynamic risk framework for urban built environments that tracks flood and earthquake exposure together with temporal changes in use and vulnerability.
30. Developing Community Disaster Resilience Through Collaborative Professional Development: Integrating Teachers, Informal Educators, and Emergency Management Personnel
Core Problem: Community disaster resilience depends on preparedness networks that extend beyond emergency managers alone, yet those educational linkages are rarely formalized.
Key Innovation: This work shows how collaborative professional development across teachers, informal educators, and emergency management personnel can strengthen community-scale disaster resilience.