Initiated by Dr. Xin Wei, University of Michigan
Ongoing development by the community

TerraMosaic Daily Digest: May 28, 2026

Daily Summary

May 28 advances hazard science by tightening the link between observation, process, and decision geometry. Coastal cliff runout is measured as an empirical distribution rather than inferred from generic setbacks, while TSUSY turns historical tsunami occurrence into a traceable global data product. Thawing moraines, lava-delta retreat, rock-glacier deactivation, layered-loess erosion, and soil-rock mixture slopes all show terrain instability as a coupled response to thermal, hydrologic, structural, or material heterogeneity. The central advance is the conversion of landform change, event history, and deformation into process constraints that can be used in hazard zoning, scenario modelling, or design.

The hydrological and infrastructure papers are similarly process-centred. Flood risk is treated across mechanisms and scales, from snow-dominated catchment shifts and optical-SAR inundation mapping to neural-operator urban flood simulation and glacial-lake downstream consequences. Earthquake and underground-engineering studies emphasize friction laws, liquefaction intensity measures, spatially variable metro response, submarine-tunnel fragility, deep TBM collapse, and low-cost liquefaction mitigation. AI and remote-sensing contributions are clearest where they improve transfer under real deployment limits: multimodal change detection, distribution-shift benchmarks, onboard damage assessment, AI weather assimilation, SAR-derived 3-D displacement, and sensor-fusion methods for flood, snow, wetland, and infrastructure monitoring.

Key Trends

Five methodological movements stand out: empirical hazard-zone calibration, thermally coupled cold-region deformation, mechanism-aware flood modelling, heterogeneity-aware infrastructure risk, and deployment-tested AI for Earth observation.

  • Hazard zoning is becoming distributional and process-calibrated: The coastal landslide runout record, TSUSY tsunami database, Makran synthesis, tsunami-like scour experiments, liquefaction intensity-measure analysis, and log-pile mitigation tests all replace broad assumptions with event- or mechanism-specific evidence.
  • Cold-region and mountain hazards are being framed as coupled thermal-hydrologic deformation problems: Thaw-induced moraine motion, Dry Andes rock-glacier deactivation, permafrost ground-temperature prediction, freeze-thaw seepage in loess, snow-dominated flood-mechanism shifts, duct-ventilated embankment cooling, and Tibetan Plateau lake dynamics connect climate forcing to terrain or infrastructure response.
  • Flood science is linking physical mechanisms, observation fusion, and fast simulation: The Miyun optical-SAR flood map, LarNO urban flood model, snow-catchment flood-mechanism study, endorheic flood mapping, glacial-lake hydrology, Lake Eyre SWOT analysis, and long-term lake-storage work show a movement from flood extent products toward state-aware flood dynamics.
  • Infrastructure geohazard assessment is explicitly incorporating spatial variability and system resilience: Soil-rock mixture slope stability, metro-station seismic prediction, submarine-tunnel resilience, deep hard-rock TBM collapse, tunnel digital twins, reclaimed-ground embankment response, offshore scour, and CPT inversion all treat material heterogeneity or system fragility as central design variables.
  • AI for geoscience is being tied to sensor constraints, physics, and deployment shift: OmniCD, EarthShift, onboard building-damage assessment, AI weather data assimilation, long-rollout AI forecast benchmarking, Capella 3-D displacement retrieval, temporal satellite reconstruction, semi-supervised change detection, and optical or SAR fusion methods share a common requirement: they test transferability and observation limits rather than only benchmark accuracy.

Selected Papers

This issue contains 70 selected papers from 1,770 papers analyzed. The papers are anchored by empirical coastal landslide runout, a global tsunami-event database, thaw-induced moraine deformation, optical-SAR flood mapping, neural-operator urban flood modelling, changing snow-dominated flood mechanisms, earthquake friction laws, Makran tsunami synthesis, spatially variable soil-rock mixture slope stability, deep TBM tunnel collapse, layered-loess erosion, and liquefaction mitigation. The broader set extends those themes through lava-delta retreat, rock-glacier deactivation, tsunami-like scour, permafrost temperature prediction, metro and submarine-tunnel seismic resilience, SAR-derived 3-D displacement, building-damage and gully-erosion mapping, glacial-lake hydrology, compound drought, AI weather assimilation, Earth-observation robustness, and heterogeneity-aware geotechnical design.

1. Observations of coastal cliff landslide runout in southern California from 21 years of data

Source: Earth Surface Processes and Landforms Type: Coastal Landslide Runout Observation and Hazard-Zone Calibration Geohazard Type: Coastal cliff landslides, beach runout, cliff-retreat hazard, LiDAR and photogrammetric time series Relevance: 9/10

Core Problem: Coastal-cliff hazard zones often rely on generic setbacks even though debris runout on beaches is controlled by local cliff geometry and failure history.

Key Innovation: Earth Surface Processes and Landforms compiles 21 years and more than 700 surveys from southern California, quantifying runout distributions and showing that one-half cliff height captures nearly all observed failures.

2. The TSUSY Database: a global database of historical tsunami events and a tsunami-occurrence criterion based on historical earthquakes

Source: Natural Hazards and Earth System Sciences Type: Global Historical Tsunami Database and Occurrence Criterion Geohazard Type: Tsunamis, tsunamigenic earthquakes, global event catalogues, source-to-impact hazard screening Relevance: 9/10

Core Problem: Tsunami hazard analysis needs a consistent global record of events and a defensible criterion for linking historical earthquakes to tsunami occurrence.

Key Innovation: Natural Hazards and Earth System Sciences presents TSUSY, a global historical tsunami database, and defines an earthquake-based occurrence criterion for reproducible event classification.

3. Deciphering thaw-induced deformation signatures of ice-rich moraines and periglacial permafrost using MT-InSAR

Source: Geomorphology Type: MT-InSAR Monitoring of Thaw-Induced Moraine Deformation Geohazard Type: Ice-rich moraines, periglacial permafrost, thaw deformation, alpine slope instability, MT-InSAR Relevance: 8/10

Core Problem: Ice-rich moraines can deform under thawing permafrost before overt failure, but their spatially variable signatures are difficult to isolate with sparse field data.

Key Innovation: The study uses multi-temporal InSAR to decipher thaw-induced deformation patterns in periglacial moraines, connecting surface motion to permafrost degradation and moraine stability.

4. Machine learning–based flood inundation mapping using fused optical and SAR remote sensing: a case study of the Miyun flood

Source: Frontiers in Earth Science Type: Optical-SAR Machine Learning for Flood Inundation Mapping Geohazard Type: Flood inundation, optical-SAR fusion, remote sensing, machine learning, Miyun flood case study Relevance: 8/10

Core Problem: Flood mapping during cloudy or rapidly evolving events is limited when optical and SAR observations are used separately.

Key Innovation: The paper fuses optical and SAR remote-sensing data in a machine-learning workflow to delineate flood inundation for the Miyun flood, improving event-scale water mapping under mixed sensor constraints.

5. Large-scale urban flood modeling and zero-shot high-resolution generalization with LarNO

Source: Journal of Hydrology Type: Neural-Operator Urban Flood Modelling with Zero-Shot Generalization Geohazard Type: Urban flooding, large-scale hydrodynamic modelling, neural operators, high-resolution generalization Relevance: 8/10

Core Problem: High-resolution urban flood simulations are computationally expensive and hard to transfer across scales or unseen domains.

Key Innovation: LarNO uses a neural-operator architecture for large-scale urban flood modelling and demonstrates zero-shot high-resolution generalization, pointing toward rapid city-scale flood scenario generation.

6. Changing Flood‐Generating Mechanisms Impact Flood Characteristics in Snow‐Dominated Catchments

Source: Geophysical Research Letters Type: Mechanistic Change in Snow-Dominated Flood Generation Geohazard Type: Snowmelt floods, rain-on-snow, catchment hydrology, changing flood characteristics, climate sensitivity Relevance: 8/10

Core Problem: Flood frequency and magnitude in snow-dominated catchments can change because the dominant generating mechanism shifts, not only because precipitation totals change.

Key Innovation: Geophysical Research Letters links changing flood characteristics to altered flood-generating mechanisms, separating snowmelt, rainfall, and mixed-process contributions in cold-region hydrology.

7. Multi‐Scale Rate‐ and Roughness‐Dependent Frictional Constitutive Law and Dynamic Earthquake Sequence Simulation

Source: Journal of Geophysical Research: Solid Earth Type: Rate- and Roughness-Dependent Earthquake Friction Law Geohazard Type: Earthquake sequence simulation, fault friction, dynamic rupture, rate-state mechanics, rough fault surfaces Relevance: 8/10

Core Problem: Earthquake-cycle models need friction laws that reproduce both rate dependence and rough-surface effects across scales.

Key Innovation: The study formulates a multi-scale rate- and roughness-dependent constitutive law and applies it to dynamic earthquake sequence simulation.

8. 80 Years of research on tsunamigenic earthquakes in the Makran subduction zone (1945–2025): a review- Part A: MSZ features

Source: Frontiers in Earth Science Type: Makran Tsunamigenic Earthquake Synthesis Geohazard Type: Makran subduction zone, tsunamigenic earthquakes, regional tsunami hazard, historical review Relevance: 8/10

Core Problem: The Makran subduction zone has high tsunami potential but uneven historical, geological, and seismological constraints.

Key Innovation: The review synthesizes 80 years of research on Makran tsunamigenic earthquakes, consolidating source-zone features relevant to regional tsunami hazard assessment.

9. Stability assessment of SRM slopes considering spatial randomness and cross-correlated spatial variability via a random NCDDAM

Source: Journal of Rock Mechanics and Geotechnical Engineering Type: Spatial-Randomness Stability Assessment of Soil-Rock Mixture Slopes Geohazard Type: Soil-rock mixture slopes, spatial variability, cross-correlation, slope stability, random NCDDAM Relevance: 8/10

Core Problem: Slope stability in soil-rock mixtures depends on spatially variable and cross-correlated material properties that deterministic analyses often suppress.

Key Innovation: The paper evaluates SRM slope stability with a random NCDDAM framework that represents spatial randomness and cross-correlated variability.

10. Multi-stage process of stress-structure-induced collapse in a deep hard-rock TBM tunnel

Source: Journal of Rock Mechanics and Geotechnical Engineering Type: Stress-Structure-Induced Collapse in Deep Hard-Rock TBM Tunnelling Geohazard Type: Tunnel collapse, deep hard rock, TBM excavation, stress-structure interaction, underground geohazards Relevance: 8/10

Core Problem: Deep TBM tunnels can collapse through staged interaction between high in situ stress and adverse rock structure, but the process is difficult to reconstruct.

Key Innovation: The study resolves a multi-stage stress-structure-induced collapse mechanism in a deep hard-rock TBM tunnel, linking excavation response to structural controls.

11. Soil erosion processes on layered loess slopes under extreme rainfall events

Source: Journal of Hydrology Type: Extreme-Rainfall Erosion Processes on Layered Loess Slopes Geohazard Type: Loess slope erosion, extreme rainfall, layered soils, hillslope sediment transport, shallow failure precursors Relevance: 8/10

Core Problem: Layered loess slopes can erode and destabilize rapidly during extreme rainfall, yet the process sequence is sensitive to stratigraphy.

Key Innovation: The paper characterizes soil erosion processes on layered loess slopes under extreme rainfall, clarifying how stratification modulates runoff, erosion, and slope degradation.

12. Shallow ground improvement by log piles for liquefaction mitigation of small residential buildings: experimental study by 1-g shaking table tests

Source: Soils and Foundations Type: Log-Pile Ground Improvement for Liquefaction Mitigation Geohazard Type: Soil liquefaction, residential foundations, shallow ground improvement, log piles, shaking-table tests Relevance: 8/10

Core Problem: Small residential buildings often need low-cost liquefaction mitigation methods that are practical for shallow ground improvement.

Key Innovation: One-g shaking-table experiments test log piles as shallow ground improvement, evaluating their ability to reduce liquefaction effects beneath small residential buildings.

13. Rapid changes of the lava‐delta coastlines formed by the 2021 volcanic eruption on La Palma, Canary Islands

Source: Earth Surface Processes and Landforms Type: Lava-Delta Coastline Change After the La Palma Eruption Geohazard Type: Volcanic coastal hazards, lava deltas, cliff retreat, wave erosion, high-resolution satellite monitoring Relevance: 7/10

Core Problem: New lava deltas can retreat rapidly because fresh volcanic fronts combine weak internal structure with energetic wave attack.

Key Innovation: Earth Surface Processes and Landforms uses sub-metre satellite imagery to track daily-to-monthly shoreline retreat after the 2021 La Palma eruption and links contrasting delta behaviour to emplacement structure.

14. The Lizoite rock glacier, Dry Andes of Northwestern Argentina: A deactivation case in progress?

Source: Earth Surface Processes and Landforms Type: Rock-Glacier Deactivation in the Dry Andes Geohazard Type: Rock glaciers, permafrost degradation, cryosphere geomorphology, surface velocity, warming response Relevance: 7/10

Core Problem: Dry Andes rock glaciers are widespread but poorly constrained in their thermal and kinematic transition from active to inactive states.

Key Innovation: The study combines GNSS velocity, ground-temperature monitoring, and geomorphic mapping to diagnose progressive deactivation of the Lizoite rock glacier.

15. Fluid‐Induced Magnetic Enhancement in Sandstone Friction Experiments: Implications for Coseismic Fault Temperature Estimates

Source: Geophysical Research Letters Type: Magnetic Enhancement During Sandstone Friction Experiments Geohazard Type: Fault friction, coseismic temperature estimates, magnetic mineral alteration, earthquake mechanics Relevance: 7/10

Core Problem: Magnetic proxies for coseismic heating can be altered by fluids and frictional processes, complicating earthquake temperature estimates.

Key Innovation: The experiments show fluid-induced magnetic enhancement in sandstone friction tests, refining interpretation of magnetic signals in fault-heating studies.

16. Impoundment Depth Effects on Bore Hydrodynamics and Scour around a Berm-Mounted Structure during Tsunami-Like Inundation and Drawdown

Source: Coastal Engineering Type: Tsunami-Like Bore Hydrodynamics and Structure Scour Geohazard Type: Tsunami inundation, drawdown, coastal scour, berm-mounted structures, hydraulic experiments Relevance: 7/10

Core Problem: Scour around coastal structures during tsunami inundation and drawdown depends on bore depth and flow reversal, but the coupled response is difficult to parameterize.

Key Innovation: The study quantifies impoundment-depth effects on bore hydrodynamics and scour around a berm-mounted structure during tsunami-like inundation and drawdown.

17. Integrating physics-informed data augmentation and ensemble learning for ground temperature prediction in permafrost regions

Source: Cold Regions Science and Technology Type: Physics-Informed Permafrost Ground-Temperature Prediction Geohazard Type: Permafrost, ground temperature, physics-informed data augmentation, ensemble learning, cold-region infrastructure Relevance: 7/10

Core Problem: Permafrost ground-temperature prediction is limited by sparse observations and strong seasonal thermal dynamics.

Key Innovation: The paper integrates physics-informed data augmentation with ensemble learning to improve ground-temperature prediction in permafrost regions.

18. Seismic stability prediction of metro stations considering soil spatial variability: A stacked ensemble machine learning approach

Source: Tunnelling and Underground Space Technology Type: Machine-Learning Seismic Stability Prediction for Metro Stations Geohazard Type: Metro stations, seismic stability, soil spatial variability, stacked ensemble learning, urban underground risk Relevance: 7/10

Core Problem: Seismic stability of metro stations is sensitive to spatially variable soil conditions that are difficult to propagate through routine design.

Key Innovation: A stacked ensemble machine-learning model predicts metro-station seismic stability while accounting for soil spatial variability.

19. Probabilistic seismic resilience assessment integrating optimal IM selection and system-level fragility analysis: Case study of submarine tunnel

Source: Tunnelling and Underground Space Technology Type: Probabilistic Seismic Resilience of Submarine Tunnels Geohazard Type: Submarine tunnels, seismic resilience, fragility analysis, optimal intensity measures, system reliability Relevance: 7/10

Core Problem: Submarine tunnel resilience depends on choosing intensity measures that link seismic demand to system-level fragility.

Key Innovation: The study integrates optimal ground-motion intensity-measure selection with probabilistic system-level fragility analysis for a submarine tunnel case.

20. Ground Motion Intensity Measures at Liquefaction Field Case History Sites

Source: Journal of Geotechnical and Geoenvironmental Engineering Type: Ground-Motion Intensity Measures at Liquefaction Case-History Sites Geohazard Type: Liquefaction, ground-motion intensity measures, field case histories, seismic hazard calibration Relevance: 7/10

Core Problem: Liquefaction triggering models depend on the intensity measures assigned to historical field case histories.

Key Innovation: The paper evaluates ground-motion intensity measures at liquefaction field case-history sites, supporting better calibration of empirical liquefaction assessment.

21. Mapping Flood in Endorheic Depressions Using Multitemporal and Multiresolution Remote Sensing Data—Example of Chotts Merouane and Melrhir, Algeria

Source: GeoHazards Type: Remote-Sensing Flood Mapping in Endorheic Depressions Geohazard Type: Flooding, endorheic depressions, multitemporal remote sensing, Chotts Merouane and Melrhir, Algeria Relevance: 7/10

Core Problem: Flood mapping in closed basins is complicated by shallow water, changing salinity, and intermittent inundation.

Key Innovation: GeoHazards maps flooding in Algerian endorheic depressions using multitemporal and multiresolution remote-sensing data.

22. Joint Hail Detection from Satellite and Radar Observations with Spatially Adaptive Alignment and Wavelet-Gated Refinement

Source: Remote Sensing Type: Joint Satellite-Radar Hail Detection Geohazard Type: Hail hazards, severe convective storms, satellite-radar fusion, spatial alignment, wavelet refinement Relevance: 7/10

Core Problem: Hail detection needs to reconcile satellite context with radar structure under spatial misalignment and multi-scale storm texture.

Key Innovation: Remote Sensing combines satellite and radar observations with spatially adaptive alignment and wavelet-gated refinement for hail detection.

23. Retrieval of 3-D Ground Displacement Time Series From Multitemporal/Multiangle Capella Space SAR Data Acquired From Mid-Inclination Orbits

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: Three-Dimensional Ground Displacement from Capella SAR Geohazard Type: Ground deformation, 3-D displacement, multitemporal SAR, multiangle observations, infrastructure monitoring Relevance: 7/10

Core Problem: Single-view SAR deformation products cannot fully resolve three-dimensional ground motion.

Key Innovation: The study retrieves 3-D ground-displacement time series from multitemporal and multiangle Capella Space SAR data acquired from mid-inclination orbits.

24. DSDRN: A Dual-Stream Dual-Refinement Network With Superpixel and Location Semantics for Building Damage Classification

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: Dual-Stream Network for Building Damage Classification Geohazard Type: Building damage, disaster remote sensing, superpixels, location semantics, post-event assessment Relevance: 7/10

Core Problem: Post-disaster building-damage mapping must preserve both object boundaries and spatial context under heterogeneous damage patterns.

Key Innovation: DSDRN combines superpixel and location semantics in a dual-stream dual-refinement network for building damage classification.

25. A Hybrid CNN-Transformer Architecture for Gully Erosion Extraction in Northeast China Using High-Resolution Images

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: CNN-Transformer Gully Erosion Extraction Geohazard Type: Gully erosion, high-resolution imagery, geomorphic mapping, CNN-transformer models, Northeast China Relevance: 7/10

Core Problem: Gully erosion mapping from high-resolution imagery requires both local texture recognition and longer-range spatial context.

Key Innovation: The paper develops a hybrid CNN-transformer architecture to extract gully erosion features in Northeast China.

26. Downstream hydrological consequences of existing and potential glacial lakes in a changing climate

Source: Journal of Hydrology Type: Hydrological Consequences of Existing and Potential Glacial Lakes Geohazard Type: Glacial lakes, downstream hydrology, climate change, mountain water hazards, future lake development Relevance: 7/10

Core Problem: Expanding glacial lakes can alter downstream hydrology as well as outburst-flood exposure.

Key Innovation: Journal of Hydrology quantifies downstream hydrological consequences of existing and potential glacial lakes under a changing climate.

27. Seepage evolution in loess driven by freeze-thaw cycling: a multiscale investigation

Source: Journal of Hydrology Type: Freeze-Thaw Driven Seepage Evolution in Loess Geohazard Type: Loess, freeze-thaw cycling, seepage, multiscale pore change, cold-region slope hydrology Relevance: 7/10

Core Problem: Freeze-thaw cycling can restructure loess pore networks and change seepage pathways that influence slope stability.

Key Innovation: The paper investigates seepage evolution in loess under freeze-thaw cycling using a multiscale framework.

28. Land use and land cover change have reduced meteorological-hydrological compound drought on a global scale

Source: Journal of Hydrology Type: Land-Cover Effects on Compound Drought Geohazard Type: Meteorological-hydrological compound drought, land-cover change, global hydrology, drought risk Relevance: 7/10

Core Problem: Compound drought risk reflects both climate forcing and land-cover effects on water partitioning.

Key Innovation: Journal of Hydrology reports that land use and land-cover change have reduced meteorological-hydrological compound drought on a global scale.

29. Ensemble‐Based Assimilation of Sounding Observations With AI Weather Models

Source: Geophysical Research Letters Type: Sounding-Observation Assimilation in AI Weather Models Geohazard Type: AI weather models, data assimilation, sounding observations, extreme-weather prediction, forecast initialization Relevance: 7/10

Core Problem: AI weather forecasts need practical data-assimilation strategies that can ingest sparse but information-rich atmospheric soundings.

Key Innovation: Geophysical Research Letters develops ensemble-based assimilation of sounding observations for AI weather models.

30. OmniCD: A Foundational Framework for Remote Sensing Image Change Detection Guided by Multimodal Semantics

Source: ArXiv (Geo/RS/AI) Type: Multimodal Foundation Framework for Remote-Sensing Change Detection Geohazard Type: Remote-sensing change detection, multimodal semantics, foundation models, disaster mapping transfer Relevance: 7/10

Core Problem: Change-detection systems often fail when visual appearance shifts across sensors, locations, or event types.

Key Innovation: OmniCD introduces a foundational remote-sensing change-detection framework guided by multimodal semantics.

31. EarthShift: a benchmark for measuring robustness to real-world distribution shifts in Earth observation

Source: ArXiv (Geo/RS/AI) Type: Earth Observation Robustness Benchmark Under Distribution Shift Geohazard Type: Earth observation, distribution shift, model robustness, benchmark datasets, disaster-transfer evaluation Relevance: 7/10

Core Problem: Remote-sensing models used in hazards can degrade sharply when deployed under real-world distribution shifts.

Key Innovation: EarthShift provides a benchmark for measuring robustness to real-world distribution shifts in Earth observation.

32. Can AI Weather Models Predict Beyond Two Weeks? A Quantitative Benchmark and Analysis of Long Rollouts

Source: ArXiv (Geo/RS/AI) Type: Long-Rollout Benchmark for AI Weather Models Geohazard Type: AI weather prediction, long-range forecast skill, benchmark evaluation, hazard lead time Relevance: 7/10

Core Problem: Hazard applications need to know where AI weather models lose skill beyond the short-range window.

Key Innovation: The benchmark quantifies whether AI weather models can predict beyond two weeks and analyzes long-rollout degradation.

33. Optimizing Latent Representations for Robust Building Damage Assessment Onboard Earth Observation Satellites

Source: ArXiv (Geo/RS/AI) Type: Onboard Building-Damage Assessment from Earth Observation Satellites Geohazard Type: Building damage, onboard satellite processing, latent representations, rapid disaster response, edge inference Relevance: 7/10

Core Problem: Disaster response can be delayed when raw satellite imagery must be downlinked before damage assessment.

Key Innovation: The paper optimizes latent representations for robust building-damage assessment onboard Earth-observation satellites.

34. Virtual testbed for multi-risk assessment: defining RETURNVILLEs to support the analysis and testing of DRM and CCA solutions in realistic urban contexts

Source: International Journal of Disaster Risk Reduction Type: Virtual Urban Testbed for Multi-Risk Assessment Geohazard Type: Multi-risk assessment, urban digital testbeds, disaster risk management, climate adaptation, RETURNVILLEs Relevance: 7/10

Core Problem: Urban disaster-risk methods need realistic virtual environments where DRM and climate-adaptation solutions can be stress-tested.

Key Innovation: The paper defines RETURNVILLEs as virtual testbeds for multi-risk assessment in realistic urban contexts.

35. Disentangling Management and Climate Drivers in an Anthropogenic Transitional Mediterranean Coastal Groundwater-Dependent Ecosystem

Source: Remote Sensing Type: Remote-Sensing Attribution in a Coastal Groundwater-Dependent Ecosystem Geohazard Type: Coastal groundwater, climate and management drivers, ecosystem water stress, remote sensing, Mediterranean wetlands Relevance: 7/10

Core Problem: Coastal groundwater-dependent ecosystems respond to both management and climate forcing, which are difficult to disentangle observationally.

Key Innovation: Remote Sensing separates management and climate drivers in an anthropogenic transitional Mediterranean coastal groundwater-dependent ecosystem.

36. Reconstruction of Global 0.25° Land Lightning Density from 1979 to 2025 based on an ensemble machine learning

Source: Earth System Science Data Type: Global Ensemble-Reconstructed Lightning Density Geohazard Type: Lightning hazards, global climatology, ensemble machine learning, long-term gridded dataset, convective extremes Relevance: 7/10

Core Problem: Lightning risk analysis needs long, spatially consistent global density records that predate modern satellite coverage.

Key Innovation: The paper reconstructs global 0.25-degree land lightning density from 1979 to 2025 using ensemble machine learning.

37. Volumetric analysis of a playa lake using SWOT data: An improved understanding of the inflows to Kati Thanda-Lake Eyre

Source: Journal of Hydrology Type: SWOT-Based Volumetric Analysis of Kati Thanda-Lake Eyre Geohazard Type: Lake hydrology, SWOT, volumetric water storage, arid-zone flooding, inflow reconstruction Relevance: 7/10

Core Problem: Intermittent playa lakes are difficult to quantify volumetrically because water extent and elevation change rapidly.

Key Innovation: Journal of Hydrology uses SWOT data for volumetric analysis of Kati Thanda-Lake Eyre, improving understanding of inflows.

38. Temporal scale dictates the dominant driver of lake dynamics on the Tibetan Plateau

Source: Journal of Hydrology Type: Temporal-Scale Controls on Tibetan Plateau Lake Dynamics Geohazard Type: Lake dynamics, Tibetan Plateau, temporal-scale attribution, climate controls, high-altitude hydrology Relevance: 7/10

Core Problem: Dominant controls on lake dynamics can change with the temporal scale of analysis.

Key Innovation: The study shows that temporal scale dictates the dominant driver of lake dynamics on the Tibetan Plateau.

39. Laser bathymetry on rough riverbed channels: State‐of‐the‐art and future prospects

Source: Earth Surface Processes and Landforms Type: Laser Bathymetry for Rough Riverbed Channels Geohazard Type: Rough-bed rivers, topo-bathymetric LiDAR, UAV sensing, river morphology, sediment transport Relevance: 6/10

Core Problem: Rough, shallow, turbulent rivers remain difficult for bathymetric mapping despite their importance for sediment transport and restoration.

Key Innovation: Earth Surface Processes and Landforms reviews laser bathymetry for rough riverbeds and identifies sensor and deployment limits for future UAV-borne surveys.

40. Denudation rates of carbonate coasts: Insights from in situ cosmogenic 36CL from Cuban coastal terraces

Source: Earth Surface Processes and Landforms Type: Cosmogenic-Nuclide Denudation Rates of Carbonate Coasts Geohazard Type: Carbonate coasts, coastal terraces, hurricane-exposed erosion, denudation rates, cosmogenic nuclides Relevance: 6/10

Core Problem: Rocky carbonate coasts record both marine and continental erosion, but the relative controls on denudation vary across terrace position and age.

Key Innovation: Earth Surface Processes and Landforms combines in situ cosmogenic 36Cl with coral dating to quantify denudation rates on Cuban coastal terraces.

41. Temporal dune growth dynamics based on a 3‐year monitored dune featuring marram grass and brushwood fences

Source: Earth Surface Processes and Landforms Type: Monitored Dune Growth for Coastal Flood Protection Geohazard Type: Coastal dunes, flood protection, sand trapping, drone monitoring, adaptive management Relevance: 6/10

Core Problem: Nature-based dune interventions for flood protection require evidence on how vegetation, brushwood fences, and management choices affect sand accumulation over multiple years.

Key Innovation: Earth Surface Processes and Landforms uses three years of drone surveys to quantify dune growth and sand-trapping efficiency under different intervention designs.

42. Global Fair‐Weather Bias in Remotely Sensed Coastal Suspended Sediment Concentration

Source: Geophysical Research Letters Type: Fair-Weather Bias in Coastal Suspended-Sediment Remote Sensing Geohazard Type: Coastal sediment, satellite observation bias, storm-wave conditions, shoreline morphodynamics, hazard monitoring Relevance: 6/10

Core Problem: Satellite sediment records can under-sample stormy conditions that control coastal erosion and sediment transport.

Key Innovation: Geophysical Research Letters quantifies global fair-weather bias in remotely sensed coastal suspended-sediment concentration.

43. Highland Pathways Shape Global Dust Vertical Transport and Its Climate Effects

Source: Geophysical Research Letters Type: Highland Pathways for Global Dust Transport Geohazard Type: Dust transport, highland pathways, vertical mixing, climate effects, aeolian hazards Relevance: 6/10

Core Problem: Dust-climate effects depend on how terrain pathways inject particles vertically into the atmosphere.

Key Innovation: Geophysical Research Letters shows how highland pathways shape global dust vertical transport and climate effects.

44. An improved modelling chain for bias-adjusted high-resolution climate and hydrological projections for Norway

Source: Geoscientific Model Development Type: Bias-Adjusted Climate-Hydrology Projection Chain for Norway Geohazard Type: Climate projections, hydrological modelling, bias adjustment, flood and water-resource assessment, Norway Relevance: 6/10

Core Problem: Hydrological risk studies require climate projections that are bias-adjusted and downscaled coherently through a modelling chain.

Key Innovation: Geoscientific Model Development presents an improved high-resolution climate and hydrological projection chain for Norway.

45. Informing Thin-Layer Placement for Coastal Wetland Restoration Through Remote Sensing and Community Outreach

Source: Remote Sensing Type: Remote Sensing for Thin-Layer Coastal Wetland Restoration Geohazard Type: Coastal wetlands, thin-layer placement, restoration monitoring, remote sensing, community-informed management Relevance: 6/10

Core Problem: Wetland restoration by thin-layer sediment placement needs spatial evidence for where elevation and vegetation response justify intervention.

Key Innovation: Remote Sensing uses remote sensing and community outreach to inform thin-layer placement for coastal wetland restoration.

46. Post-Stack Seismic Inversion with Non-Convex Total Generalized Variation Regularization

Source: Remote Sensing Type: Non-Convex TGV Post-Stack Seismic Inversion Geohazard Type: Seismic inversion, subsurface imaging, non-convex regularization, geologic structure, exploration geophysics Relevance: 6/10

Core Problem: Post-stack seismic inversion must recover sharp subsurface structure while controlling noise and ill-posedness.

Key Innovation: Remote Sensing applies non-convex total generalized variation regularization to post-stack seismic inversion.

47. Lithospheric Thermal Structure Beneath East Antarctica Derived from Aeromagnetic Anomaly Analysis

Source: Remote Sensing Type: Aeromagnetic Thermal Structure Beneath East Antarctica Geohazard Type: East Antarctica, lithospheric thermal structure, aeromagnetic anomalies, geothermal boundary conditions, ice-sheet modelling Relevance: 6/10

Core Problem: Ice-sheet and geodynamic models depend on lithospheric thermal structure that is difficult to measure beneath East Antarctica.

Key Innovation: Remote Sensing derives lithospheric thermal structure beneath East Antarctica from aeromagnetic anomaly analysis.

48. CMF-SnowNet: Cross-Modal Fusion With Attent-Guided GAN and Temporal ConvNeXt for Long-Term Snow Cover Prediction

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: Cross-Modal Snow-Cover Prediction Network Geohazard Type: Snow cover, multimodal remote sensing, GAN fusion, temporal ConvNeXt, cryosphere monitoring Relevance: 6/10

Core Problem: Long-term snow-cover prediction requires combining complementary sensors across cloudy and data-limited periods.

Key Innovation: CMF-SnowNet uses cross-modal fusion with an attention-guided GAN and temporal ConvNeXt for long-term snow-cover prediction.

49. Point-DCA-Enhanced Deep Learning Network for Photon Signal Classification in UAV-Based Photon-Counting LiDAR Bathymetry

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: Photon-Counting UAV LiDAR Bathymetry Classification Geohazard Type: UAV bathymetry, photon-counting LiDAR, river and coastal mapping, point-cloud classification Relevance: 6/10

Core Problem: Photon-counting bathymetric LiDAR needs robust separation of signal and noise in shallow-water UAV surveys.

Key Innovation: The paper introduces a Point-DCA-enhanced deep network for photon signal classification in UAV-based photon-counting LiDAR bathymetry.

50. A Bézier Curve-Based Progressive TIN Densification Filtering Algorithm for Airborne LiDAR Data in Complex Transmission Corridor Terrains

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: LiDAR Terrain Filtering in Transmission Corridors Geohazard Type: Airborne LiDAR, terrain filtering, transmission corridors, complex terrain, infrastructure exposure mapping Relevance: 6/10

Core Problem: Terrain extraction in transmission corridors is difficult where vegetation, infrastructure, and complex relief overlap.

Key Innovation: The study proposes a Bezier-curve progressive TIN densification filtering algorithm for airborne LiDAR in complex transmission-corridor terrain.

51. Miniaturized Coherent Doppler Wind Lidar with Self-Compensating Harris Hawks Optimization Algorithm for Low-Altitude UAV-Borne Wind Sensing

Source: Remote Sensing Type: Miniaturized UAV Doppler Wind LiDAR Geohazard Type: UAV wind sensing, coherent Doppler LiDAR, low-altitude atmospheric monitoring, optimization algorithms Relevance: 6/10

Core Problem: Low-altitude wind monitoring for hazards and boundary-layer studies requires compact sensors with stable self-compensation.

Key Innovation: Remote Sensing develops a miniaturized coherent Doppler wind LiDAR with self-compensating Harris Hawks optimization for UAV deployment.

52. Semi-Supervised Change Detection for High-Resolution Remote Sensing Images Based on Label Extension

Source: Remote Sensing Type: Semi-Supervised High-Resolution Remote-Sensing Change Detection Geohazard Type: Change detection, high-resolution imagery, semi-supervised learning, label extension, disaster mapping Relevance: 6/10

Core Problem: High-resolution change detection is limited by scarce labelled data across different scenes and disturbance types.

Key Innovation: Remote Sensing proposes a semi-supervised change-detection method based on label extension.

53. A Review on Super-Resolution Reconstruction of Single-Frame Remote Sensing Images via Diffusion Models

Source: Remote Sensing Type: Diffusion-Based Remote-Sensing Super-Resolution Review Geohazard Type: Remote-sensing super-resolution, diffusion models, image reconstruction, observation enhancement Relevance: 6/10

Core Problem: Single-frame remote-sensing imagery often lacks the resolution required for fine-scale hazard interpretation.

Key Innovation: Remote Sensing reviews diffusion-model approaches for single-frame remote-sensing super-resolution reconstruction.

54. Efficient spectral-temporal reconstruction of long-term satellite time series via temporal segments and mask-informed embedding

Source: Remote Sensing of Environment Type: Long-Term Satellite Time-Series Reconstruction Geohazard Type: Satellite time series, spectral-temporal reconstruction, mask-informed embedding, observation continuity Relevance: 6/10

Core Problem: Long-term satellite records contain gaps and corrupted observations that reduce their value for change analysis.

Key Innovation: The study reconstructs spectral-temporal satellite time series using temporal segments and mask-informed embedding.

55. Quantifying facility-scale CO2 emissions using spaceborne hyperspectral imageries

Source: Remote Sensing of Environment Type: Facility-Scale CO2 Emissions from Spaceborne Hyperspectral Imagery Geohazard Type: Hyperspectral satellites, facility emissions, atmospheric plumes, climate monitoring, remote sensing Relevance: 6/10

Core Problem: Facility-scale CO2 monitoring from space requires resolving small atmospheric plumes in hyperspectral observations.

Key Innovation: The paper quantifies facility-scale CO2 emissions using spaceborne hyperspectral imagery.

56. Greening-induced soil drying mitigated by climate change in China's Loess Plateau

Source: Catena Type: Greening-Induced Soil Drying on the Loess Plateau Geohazard Type: Loess Plateau, soil moisture, vegetation greening, climate change, land-surface hydrology Relevance: 6/10

Core Problem: Vegetation greening can dry soils and alter slope hydrology, but climate change can partly offset that effect.

Key Innovation: The study shows that climate change mitigates greening-induced soil drying in Chinas Loess Plateau.

57. Analysis of the cooling effect of duct-ventilated embankments with different duct spacings in high-temperature permafrost regions

Source: Cold Regions Science and Technology Type: Cooling Performance of Duct-Ventilated Permafrost Embankments Geohazard Type: Permafrost embankments, duct ventilation, thermal stabilization, transportation infrastructure Relevance: 6/10

Core Problem: High-temperature permafrost embankments require passive cooling designs that remain effective under warming.

Key Innovation: Cold Regions Science and Technology analyzes the cooling effect of duct-ventilated embankments with different duct spacings.

58. Scales and irregularities of columnar jointed rock masses: Insights from field investigations and thermo-mechanical coupling numerical simulations

Source: Engineering Geology Type: Thermo-Mechanical Modelling of Columnar Jointed Rock Masses Geohazard Type: Columnar jointed rock, scale irregularity, thermo-mechanical coupling, rock-mass behaviour Relevance: 6/10

Core Problem: Columnar jointed rock masses have scale-dependent irregularities that affect mechanical response under thermal and stress loading.

Key Innovation: The paper combines field investigation and thermo-mechanical numerical simulation to assess scales and irregularities in columnar jointed rock masses.

59. Intrusion and erosion characteristics of calcium bentonite-based engineered barrier components

Source: Engineering Geology Type: Bentonite Barrier Intrusion and Erosion Geohazard Type: Bentonite barriers, engineered containment, erosion, groundwater interaction, repository safety Relevance: 6/10

Core Problem: Engineered bentonite barriers can degrade through intrusion and erosion when interacting with groundwater and adjacent materials.

Key Innovation: The study examines intrusion and erosion characteristics of calcium bentonite-based engineered barrier components.

60. Mechanical degradation of bedded salt rock induced by ScCO2– brine exposure: Insights from nanoindentation tests

Source: Engineering Geology Type: Salt-Rock Degradation Under ScCO2-Brine Exposure Geohazard Type: Bedded salt rock, CO2 storage, brine exposure, nanoindentation, caprock integrity Relevance: 6/10

Core Problem: Subsurface CO2 storage can alter salt-rock mechanical properties through coupled ScCO2-brine exposure.

Key Innovation: The paper uses nanoindentation tests to quantify mechanical degradation of bedded salt rock induced by ScCO2-brine exposure.

61. A Cone-Penetration-Test Inversion Model Developed via Machine Learning of Physical and Virtual Calibration-Chamber Experiments

Source: Journal of Geotechnical and Geoenvironmental Engineering Type: Machine-Learned CPT Inversion from Physical and Virtual Calibration Chambers Geohazard Type: Cone penetration testing, soil-parameter inversion, calibration chambers, machine learning, site characterization Relevance: 6/10

Core Problem: CPT-based soil characterization needs inversion models that bridge laboratory calibration and virtual numerical experiments.

Key Innovation: The study develops a machine-learning CPT inversion model trained on physical and virtual calibration-chamber experiments.

62. Laser-assisted TBM disc cutter rock-breaking: Insights from grain model and thermomechanical damage coupling

Source: Tunnelling and Underground Space Technology Type: Laser-Assisted TBM Disc-Cutter Rock Breaking Geohazard Type: TBM excavation, rock breaking, laser assistance, thermomechanical damage, grain-scale modelling Relevance: 6/10

Core Problem: Hard-rock TBM efficiency can be limited by high cutter forces and insufficient understanding of thermally assisted damage.

Key Innovation: The paper studies laser-assisted TBM disc-cutter rock breaking using a grain model with thermomechanical damage coupling.

63. An integrated LiDAR-Inertial-Visual framework for tunnel digital twins under construction: 3D reconstruction, centerline sampling, and point cloud semantic segmentation

Source: Tunnelling and Underground Space Technology Type: LiDAR-Inertial-Visual Tunnel Digital Twins Geohazard Type: Tunnel construction, digital twins, 3-D reconstruction, semantic point clouds, underground monitoring Relevance: 6/10

Core Problem: Tunnel digital twins require robust reconstruction and semantic segmentation under construction conditions.

Key Innovation: The paper integrates LiDAR, inertial, and visual sensing for tunnel 3-D reconstruction, centerline sampling, and point-cloud semantic segmentation.

64. Investigation of the contact pressure distribution between TBM disc cutter and rock surface using the 3D discrete element method

Source: Tunnelling and Underground Space Technology Type: DEM Contact Pressure of TBM Disc Cutters Geohazard Type: TBM disc cutters, contact pressure, discrete-element modelling, rock-tool interaction Relevance: 6/10

Core Problem: Disc-cutter performance depends on the evolving contact-pressure distribution at the cutter-rock interface.

Key Innovation: The study uses 3-D discrete-element modelling to investigate contact pressure between TBM disc cutters and rock surfaces.

65. Effects of spatial variations in physical properties of reclaimed ground consisting of rock debris on seismic behavior of embankment

Source: Soils and Foundations Type: Spatial Variability in Reclaimed Rock-Debris Embankments Geohazard Type: Reclaimed ground, rock debris, seismic embankment behaviour, spatial variability, geotechnical risk Relevance: 6/10

Core Problem: Reclaimed rock-debris embankments can respond unevenly to seismic loading because physical properties vary spatially.

Key Innovation: The paper evaluates how spatial variations in reclaimed-ground properties affect seismic behaviour of embankments.

66. Weakening Behavior of Sandstone Subjected to Microwave Irradiation: Insights from Real-Time Monitoring of Surface Temperature and Acoustic Emission

Source: Journal of Rock Mechanics and Geotechnical Engineering Type: Microwave Weakening of Sandstone Geohazard Type: Microwave rock weakening, sandstone, acoustic emission, thermal damage, excavation mechanics Relevance: 6/10

Core Problem: Microwave preconditioning can weaken rock for excavation, but damage evolution must be tracked in real time.

Key Innovation: The study monitors surface temperature and acoustic emission to characterize sandstone weakening under microwave irradiation.

67. Real-time monitoring of micro-deformation and fluid distribution during displacement in reservoirs using distributed fiber optic sensing and X-ray CT

Source: Journal of Rock Mechanics and Geotechnical Engineering Type: Fiber-Optic and X-Ray CT Reservoir Micro-Deformation Monitoring Geohazard Type: Reservoir deformation, fluid distribution, distributed fiber-optic sensing, X-ray CT, subsurface monitoring Relevance: 6/10

Core Problem: Reservoir displacement processes involve coupled micro-deformation and fluid redistribution that are hard to observe directly.

Key Innovation: The paper combines distributed fiber-optic sensing and X-ray CT for real-time monitoring of micro-deformation and fluid distribution.

68. Soil stabilization using biological CO2 sequestration and CO2-based biocementation technique

Source: Journal of Rock Mechanics and Geotechnical Engineering Type: CO2-Based Biocementation for Soil Stabilization Geohazard Type: Soil stabilization, biocementation, biological CO2 sequestration, ground improvement, carbon-aware geotechnics Relevance: 6/10

Core Problem: Ground improvement methods need to improve soil strength while reducing environmental footprint.

Key Innovation: The paper studies soil stabilization using biological CO2 sequestration and CO2-based biocementation.

69. Coarse-to-Fine Domain Incremental Learning with Attentive Distillation for Mining Footprint Segmentation in Multispectral Imagery

Source: ArXiv (Geo/RS/AI) Type: Domain-Incremental Mining Footprint Segmentation Geohazard Type: Mining footprint mapping, multispectral imagery, domain incremental learning, land-surface disturbance Relevance: 6/10

Core Problem: Mining footprint segmentation must adapt to new domains without forgetting earlier terrain and sensor conditions.

Key Innovation: The paper uses coarse-to-fine domain incremental learning with attentive distillation for mining-footprint segmentation in multispectral imagery.

70. A Multi-Agent Feedback System for Detecting and Describing News Events in Satellite Imagery

Source: ArXiv (Geo/RS/AI) Type: Multi-Agent Satellite Event Detection and Captioning Geohazard Type: Satellite event detection, disaster description, multi-agent feedback, remote-sensing interpretation Relevance: 6/10

Core Problem: Satellite imagery can show hazardous events, but turning visual evidence into reliable event descriptions remains difficult.

Key Innovation: The paper proposes a multi-agent feedback system for detecting and describing news events in satellite imagery.