TerraMosaic Daily Digest: May 25, 2026
Daily Summary
May 25 is anchored by landslide science but widens into the coupled mechanics of hazard exposure. The strongest papers quantify global flood-landslide population exposure under future climate and demographic change, replace empirical rainfall-threshold dependence with slope-unit physical prediction, and make probabilistic rainfall-landslide assessment tractable despite complex hydrological simulations. A parallel geotechnical thread resolves the materials and structures that turn forcing into failure: subaqueous slope resonance, random-field structural-jointed rock slopes, foreshock-conditioned liquefaction, internal erosion intensified by shearing, permafrost embankment cracking, and cyclic wet-dry degradation of subgrades.
The measurement papers are strongest when they create operational observables rather than generic maps. A century-scale tropical-cyclone track dataset, sixty-day streamflow forecasting, rainfall-process fingerprints, drought-propagation timing, Landsat river-mouth water color, L-band frozen-ground emission, GNSS-R wave height, Sentinel-1 wave spectra, and weakly supervised 10 m land cover all convert hard-to-observe states into quantities that can enter hazard models. The AI contribution is consequently pragmatic: Mamba segmentation, AlphaEarth land-cover distillation, multimodal image matching, fire-state inversion, and physics-guided fire-temperature reconstruction are useful because they reduce annotation burden, sensor failure, or domain shift in real environmental and infrastructure settings.
Key Trends
The papers converge on five methodological moves: replacing inventory dependence with physical prediction, coupling soil-water-structure failure pathways, resolving cold-region deformation, extending hydrometeorological observables, and using AI to overcome data scarcity or domain shift.
- Landslide assessment is moving from inventories toward physically constrained exposure modelling: The rainfall-landslide papers combine global susceptibility, future population exposure, slope-unit hydrology, threshold databases, pore-pressure simulations, and efficient reliability analysis, reducing dependence on dense historical inventories.
- Failure mechanics are treated as coupled soil-water-structure problems: Subaqueous slopes, structural-jointed rock slopes, liquefiable sands, suffusive soils, pit-in-pit excavations, unsaturated shafts, tunnel-fault interactions, and rockfill-dam materials are analyzed through interacting hydraulic, seismic, thermal, and mechanical pathways.
- Cold-region infrastructure is framed as a thermo-hydro-mechanical system: Permafrost pavement embankments, thaw-strain models, frozen peat thermal properties, freeze-thaw silt stabilization, and red-clay or mudstone railway fills link phase change, suction, particle breakage, and crack evolution to infrastructure performance.
- Hydrometeorological observables are becoming longer, finer, and more process-aware: Century-scale tropical-cyclone tracks, sixty-day streamflow warnings, drought-propagation timing, rainfall fingerprints, global river-mouth water color, and wave-height retrievals convert climate and hydrologic variability into hazard-relevant variables.
- Remote-sensing AI is most useful when it solves data bottlenecks: AlphaEarth-based land-cover distillation, Mamba segmentation, multimodal matching, LST uncertainty, frozen-ground microwave modelling, and physics-guided tunnel-fire reconstruction address label noise, sensor gaps, domain shift, and physical plausibility.
Selected Papers
This issue contains 66 selected papers from 2,304 papers analyzed. The leading papers quantify flood-landslide population exposure, replace inventory-heavy rainfall-landslide warning with slope-unit physical prediction, accelerate probabilistic rainfall-landslide reliability analysis, and resolve slope, liquefaction, and internal-erosion mechanisms under coupled hydraulic and seismic forcing. The wider set links permafrost infrastructure, subgrade degradation, coastal erosion, tunnel-fault interaction, tropical-cyclone tracks, drought propagation, streamflow warning, urban rainfall fingerprints, river-mouth water color, microwave frozen-ground retrieval, wave monitoring, remote-sensing AI, tunnel fire inversion, and lifeline vulnerability into datasets or models that can be used directly in hazard assessment.
1. Contributions of climate and population change to global flood–landslide multi-hazard population exposure under future climate scenarios
Core Problem: Flood and landslide hazards often overlap, but global exposure studies rarely separate the effects of climate change from population redistribution.
Key Innovation: Geoscience Frontiers integrates remote-sensing datasets, machine-learning susceptibility models, climate scenarios, and population projections to quantify future flood-landslide multi-hazard exposure at global scale.
2. Constructing physical-based rainfall landslides prediction model: insights from rainfall threshold curves database of slope units
Core Problem: Rainfall-threshold warnings depend on historical inventories and physical models can be too slow for regional warning when data are sparse.
Key Innovation: Natural Hazards and Earth System Sciences combines slope-unit hydrology, physical stability analysis, and threshold curves to build a faster rainfall-landslide prediction framework.
3. Efficient probabilistic assessment of rainfall-induced landslides involving complex hydrological simulations
Core Problem: Probabilistic landslide assessment is computationally expensive when full hydrological simulations and nonlinear limit-state functions are retained.
Key Innovation: JRMGE proposes a two-stage reliability approach that decouples hydraulic and mechanical uncertainties while preserving complex pore-pressure simulations for rainfall-induced landslides.
4. S-wave velocity profiles of subaqueous slopes in Lake Lucerne (Switzerland) from the inversion of full microtremor horizontal-to-vertical spectral ratios and phase velocity dispersion curves
Core Problem: Subaqueous slopes covered by unconsolidated sediment can be seismically sensitive, yet their near-surface structure is difficult to characterize without invasive surveys.
Key Innovation: Engineering Geology inverts microtremor H/V ratios and phase-velocity curves from lake OBS deployments to estimate S-wave velocity profiles for Lake Lucerne subaqueous slopes.
5. Numerical exploration of structural-jointed rock slope using a NCDDAM with random-field workflow
Core Problem: Rock-slope stability models must represent both geological structure and spatially variable jointed rock properties.
Key Innovation: Computers and Geotechnics couples node-based continuous-discontinuous deformation analysis with structural random fields to model parallel, anticlinal, and synclinal rock-slope settings.
6. Dominant mechanisms controlling sand liquefaction resistance under foreshock sequences
Core Problem: Seasonally frozen regions can experience liquefaction under coupled freeze-thaw histories and repeated seismic loading, but the controlling mechanisms remain unclear.
Key Innovation: Soil Dynamics and Earthquake Engineering uses undrained cyclic triaxial tests to isolate how foreshock sequences and freeze-thaw history alter sand liquefaction resistance.
7. Unveiling Intensified Suffusion during Triaxial Shearing via Coupled CFD-DEM Simulation
Core Problem: Internal erosion is often studied under static stress, but mechanical disturbance during shearing can intensify fine-particle loss and destabilize soil fabric.
Key Innovation: ASCE Journal of Geotechnical and Geoenvironmental Engineering uses coupled CFD-DEM with dynamic meshing and micro-CT validation to show how triaxial shearing amplifies suffusion.
8. A Reanalysis-Based Global Tropical Cyclone Tracks Dataset for the Twentieth Century (RGTracks-20C)
Core Problem: Incomplete pre-satellite tropical-cyclone records limit long-term analysis of storm exposure and historical risk.
Key Innovation: Earth System Science Data releases RGTracks-20C, a publicly available reanalysis-based global tropical cyclone track dataset spanning 1850-2014.
9. Enhancing hydrological hazard early warning: a 60 d streamflow forecasting framework integrating deep learning and process-based modeling
Core Problem: Medium-range streamflow warning remains difficult because precipitation forecast bias, catchment process simulation, and statistical post-processing must be handled together.
Key Innovation: Natural Hazards and Earth System Sciences integrates CNN bias correction, GBEHM hydrological modelling, and ARX post-processing to extend streamflow early-warning skill to 60 days.
10. Widespread acceleration of drought propagation from meteorological to ecological systems in the northern hemisphere under climate warming
Core Problem: Whether climate warming is speeding the propagation from meteorological drought to ecological drought remains unresolved across large vegetation systems.
Key Innovation: Catena combines vegetation remote sensing and meteorological data from 1982-2022 to show widespread acceleration of drought propagation across Northern Hemisphere ecosystems.
11. Unveiling urban process-dependent hydrological responses via rainfall “fingerprints”
Core Problem: Urban flood resilience depends on event-scale rainfall structure, yet rainfall heterogeneity is rarely translated into operational hydrological process classes.
Key Innovation: Journal of Hydrology extracts rainfall fingerprints from 23 years of Shanghai gauge data and links them to process-dependent urban hydrological responses.
12. Hydro-Mechanical Behavior of Sand Subgrades under Cyclic Loading across Moisture States: Insights from Large-Scale Physical Modeling
Core Problem: Flooding, groundwater rise, and moisture variation can reduce subgrade stiffness, but large-scale coupled deformation and pore-pressure datasets are scarce.
Key Innovation: Transportation Geotechnics provides large-scale physical modelling of sand subgrades across dry, saturated, compacted-moist, and water-table-controlled states under cyclic loading.
13. Mechanisms of thermo–hydro–mechanical coupling in multilayer asphalt concrete pavement embankments of permafrost regions: Triggering processes of directional deformation and cracking
Core Problem: Multilayer asphalt-concrete embankments in permafrost regions alter heat and water transport in ways that trigger directional deformation and cracking.
Key Innovation: Transportation Geotechnics investigates thermo-hydro-mechanical coupling in permafrost-region pavement embankments and identifies mechanisms that translate freezing disturbance into crack evolution.
14. Synergy of Injection Velocity, Hypergravity, and Porous Medium Skeleton for Particle Migration and Distribution Characteristics: Insights from CFD-DEM Investigations
Core Problem: Field-scale particle migration experiments are hard to interpret because pore-scale motion is hidden by medium opacity and uncertain scaling laws.
Key Innovation: ASCE Journal of Geotechnical and Geoenvironmental Engineering uses CFD-DEM to clarify how injection velocity, hypergravity, and porous skeleton geometry control particle migration.
15. Multi-Decadal Evolution Pattern and Trends of the Central Coastline of Jiangsu Province: Implications for Future Coastal Management
Core Problem: Muddy coastlines respond to both climate forcing and human development, but long-term shoreline trends require consistent multi-decadal mapping.
Key Innovation: Remote Sensing analyzes 1984-2024 imagery to quantify central Jiangsu coastline evolution and infer implications for future coastal management.
16. Dendrogeomorphic reconstruction of water erosion
Core Problem: Water erosion damages farmland, ecosystems, and infrastructure, but long-term event reconstruction remains spatially sparse.
Key Innovation: Catena reviews dendrogeomorphic theory and applications for reconstructing water erosion from tree-root and stem responses across scales.
17. Interaction mechanisms between a GIL utility tunnel and its internal structures under strike-slip faulting
Core Problem: Utility tunnels crossing strike-slip faults can experience non-uniform coupled deformation between tunnel shells and internal equipment.
Key Innovation: Tunnelling and Underground Space Technology models a fault-crossing GIL utility tunnel to resolve tunnel-internal structure interaction and the effects of deformation joints.
18. Analytical solution for kinematic response of stepped pipe shafts in unsaturated soil under vertically incident P-waves
Core Problem: Stepped pipe shafts are widely used underground, but their kinematic response under seismic P-waves in unsaturated soils is poorly constrained.
Key Innovation: Soil Dynamics and Earthquake Engineering derives an analytical solution for unsaturated soil-stepped pipe shaft interaction under vertically incident P-waves.
19. Dynamic responses of clayey gravel to fine particle loss induced by a wet/dry cycle
Core Problem: Fine-particle loss induced by wetting-drying cycles can degrade clayey gravel subgrades and alter their dynamic response.
Key Innovation: Soil Dynamics and Earthquake Engineering tests clayey gravel under coupled wet-dry and fine-loss conditions to quantify fabric degradation and cyclic deformation.
20. Mechanico-empirical model for thaw strain of fine-grained soils considering soil type and applied load
Core Problem: Thaw-induced volume change in fine-grained soils must be predicted under different soil types and loads for cold-region infrastructure design.
Key Innovation: Canadian Geotechnical Journal proposes a mechanico-empirical model for final thaw strain that accounts for soil type and applied load.
21. Thermal properties of low Arctic vegetation and peat: Coupled experimental quantification and numerical modelling for Western Siberian Lowlands
Core Problem: Permafrost peatlands depend on vegetation and peat thermal behavior, but coupled experimental and numerical estimates remain limited.
Key Innovation: Cold Regions Science and Technology quantifies low-Arctic vegetation and peat thermal properties and uses numerical modelling for Western Siberian permafrost peatlands.
22. Numerical simulation of high plasticity silt with mPCM additives subjected to freeze-thaw cycles
Core Problem: Freeze-thaw cycles degrade high-plasticity silts used in transportation infrastructure.
Key Innovation: Transportation Geotechnics combines experiments and numerical simulation to test microencapsulated phase-change materials as a stabilizer for freeze-thaw-affected silt.
23. Rate-dependent effects on mechanical properties and particle breakage of red-stratum mudstone fillers
Core Problem: Red-stratum mudstone fillers in high-fill embankments can trigger engineering distress under coupled hydro-mechanical loading.
Key Innovation: Transportation Geotechnics links strain rate, particle breakage, pore structure, and triaxial response for mudstone fillers used in eastern Tibetan embankments.
24. Performance of embankments on soft soil reinforced with rigid columns considering column-head connection type
Core Problem: Rigid column-supported embankments improve soft ground, but column-head connection effects on load transfer and failure evolution are insufficiently understood.
Key Innovation: Transportation Geotechnics compares column-head connection types through centrifuge modelling and 3D finite-element analysis of embankments on soft soil.
25. Investigation on critical speed and vibration propagation in pile-reinforced ground of ballastless high-speed railway using 2.5D FEM
Core Problem: Critical-speed and vibration propagation behavior in pile-reinforced ground remains underconstrained for ballastless high-speed railways.
Key Innovation: Transportation Geotechnics develops a coupled 2.5D FEM model with perfectly matched layers to simulate train-induced vibration in pile-reinforced ground.
26. Suction and volume evolutions of clayey soils upon high-stress unloading and subsequent soaking
Core Problem: Deep excavation failures after rainfall are partly controlled by hydro-mechanical coupling during unloading and subsequent soaking.
Key Innovation: Canadian Geotechnical Journal measures suction and volume evolution in clayey soils under high-stress unloading followed by soaking.
27. Failure mechanisms and optimization strategies for multi-level pit-in-pit excavation in soft ground
Core Problem: Multi-level pit-in-pit excavations in soft ground have complex stability controls that are not captured by simple excavation-performance analyses.
Key Innovation: Canadian Geotechnical Journal identifies failure mechanisms and optimization strategies for multi-level pit-in-pit excavation in soft ground.
28. Influence of creep behavior on cyclic cumulative strain of gravelly soil under low-frequency loading
Core Problem: Rockfill dams experience long-period water-storage cycles, but gravelly-soil strain data are mostly based on seismic-frequency loading.
Key Innovation: Soil Dynamics and Earthquake Engineering studies cyclic cumulative strain in gravelly soil under low-frequency loading relevant to reservoir operation.
29. Hydro-mechanical coupling of unsaturated red clay considering temperature and time effects
Core Problem: Unsaturated red clay used as railway subgrade fill experiences coupled effects of temperature, suction, and strain rate.
Key Innovation: JRMGE combines triaxial testing and microstructural analysis to quantify hydro-mechanical coupling in unsaturated red clay for the Sichuan-Tibet Railway.
30. Divergent pattern and trends of water color in global river mouths revealed by decadal Landsat observations
Core Problem: River mouths integrate watershed processes and human impacts, but global optical trends are poorly resolved.
Key Innovation: Journal of Hydrology uses Landsat-8 and the Forel-Ule Index to quantify water-color patterns and decadal trends across 3,209 global river mouths.
31. Canadian Wildfire Smoke Impacts on Reduced Nitrogen in the Upper Midwest: Insights From the 2023 Fire Season
Core Problem: Wildfire impacts on ammonia and ammonium deposition remain poorly quantified despite major implications for air quality and ecosystem loading.
Key Innovation: Geophysical Research Letters integrates satellite, ground, and aircraft observations to quantify reduced-nitrogen impacts from the 2023 Canadian wildfire season.
32. Improving Estimates of Prescribed Fire Smoke: Insights From a Novel Southeastern US Burn Permit Geodatabase
Core Problem: Prescribed burns reduce wildfire risk but can introduce smoke exposure, and their emissions are often poorly constrained.
Key Innovation: Geophysical Research Letters combines a southeastern U.S. burn-permit geodatabase with emissions and chemical-transport modelling to estimate prescribed-fire PM2.5 exposure.
33. Effects of Recurving Western North Pacific Tropical Cyclones on Downstream Ridges Over Pacific‐Western North America
Core Problem: Recurving western North Pacific tropical cyclones can reshape downstream ridges, but the sensitivity of circulation response to storm characteristics is not fully resolved.
Key Innovation: Geophysical Research Letters analyzes how recurving tropical cyclone characteristics affect ridge patterns over Pacific-western North America.
34. FEM‐Peridynamic Modelling of Supershear Earthquake Ruptures in Dry and Fluid‐Saturated Media
Core Problem: Supershear transition controls strong ground motion, but fluid-saturated fault behavior is difficult to reproduce in coupled rupture models.
Key Innovation: Journal of Geophysical Research uses a hybrid finite-element/peridynamic model to simulate supershear ruptures in dry and fluid-saturated media.
35. Viscosity of Trachytic Melt Constrained by Integrated Rheological and In Situ Spectroscopic Analyses: Insights From the Agnano‐Monte Spina Eruption (Campi Flegrei, Italy)
Core Problem: Trachytic magma viscosity controls ascent, degassing, and fragmentation but remains difficult to constrain under representative melt conditions.
Key Innovation: Journal of Geophysical Research combines rheology and in situ spectroscopy to develop a composition-specific viscosity model for the Agnano-Monte Spina eruption.
36. Augmented community microwave emission model to simulate multi-frequency passive microwave satellite observations over frozen ground
Core Problem: Brightness-temperature simulation over frozen and thawing ground remains difficult across sensors and frequencies.
Key Innovation: Remote Sensing of Environment augments the community microwave emission model to reproduce multi-frequency passive microwave observations over frozen ground.
37. Global Ocean Significant Wave Height Retrieval From Spaceborne GNSS-R Data Using a CNN-Transformer Network
Core Problem: Global significant wave height retrieval from GNSS-R remains limited by nonlinear relationships between reflected signals and ocean-state variables.
Key Innovation: IEEE JSTARS proposes a CNN-transformer network for global ocean significant wave height retrieval from spaceborne GNSS-R observations.
38. Distilling 10-m Land Cover Maps From Multisource Consensus via AlphaEarth Embeddings and Noise-Aware Weak Supervision
Core Problem: Scalable 10 m land-cover mapping is constrained by annotation cost and systematic label noise in existing products.
Key Innovation: IEEE JSTARS uses AlphaEarth embeddings and noise-aware weak supervision to distill 10 m land-cover maps from multisource consensus labels.
39. Between the bog and the slope: Colluvial soils of Siberian taiga and their potential for paleogeographical and geoarchaeological research
Core Problem: Colluvial soils in boreal landscapes can record erosion and land-use history, but their archive value is understudied where bog formation is extensive.
Key Innovation: Catena characterizes Siberian taiga footslope colluvial soils and evaluates their paleogeographical and geoarchaeological potential.
40. Experimental and numerical study on tunnel linings strengthened with corrugated BFRP plates
Core Problem: Tunnel linings deteriorate under groundwater, vibration, corrosion, and material aging, creating maintenance and safety challenges.
Key Innovation: Tunnelling and Underground Space Technology tests corrugated BFRP plate reinforcement for tunnel linings and validates the response with finite-element modelling.
41. A knowledge-informed data-efficient framework for soil correlation modelling
Core Problem: Geotechnical design often depends on noisy and limited soil-property datasets.
Key Innovation: Computers and Geotechnics proposes a knowledge-informed, active-learning neural framework for data-efficient soil correlation modelling.
42. A novel hybrid intelligence paradigm of ELM and enhanced slime mould algorithm based on adaptive grouping technique for predicting the bearing capacity of stiffened DCM columns
Core Problem: Bearing-capacity assessment for stiffened DCM columns remains difficult under limited service-load data.
Key Innovation: Transportation Geotechnics combines extreme learning machines with an enhanced slime-mould algorithm for predicting DCM column bearing capacity.
43. Dynamic characteristics and accumulated plastic deformation prediction of lithium slag filler for road applications based on interpretable machine learning
Core Problem: Lithium slag has potential as subgrade filler, but its accumulated plastic deformation under cyclic loading is poorly constrained.
Key Innovation: Soil Dynamics and Earthquake Engineering uses dynamic triaxial tests and interpretable ensemble machine learning to predict lithium-slag filler deformation.
44. Seismic performance of pretensioned prestressed concrete solid square piles with combined steel strand and deformed rebar
Core Problem: Traditional prestressed concrete piles can fail brittlely under low axial force ratios in high-intensity seismic regions.
Key Innovation: Soil Dynamics and Earthquake Engineering tests prestressed concrete solid square piles reinforced with steel strands and deformed rebar to improve seismic behavior.
45. Low-temperature effect on seismic life-cycle cost estimation for isolated highway bridges with high damping rubber bearings
Core Problem: High-damping rubber bearings lose effectiveness at low temperature, affecting seismic isolation performance in cold regions.
Key Innovation: Soil Dynamics and Earthquake Engineering estimates seismic life-cycle cost for isolated highway bridges under temperature-dependent bearing behavior.
46. Cumulative damage analysis and experimental investigation of transmission tower structures under sequential seismic excitations
Core Problem: Power transmission towers can accumulate damage across sequential seismic events, but experimental evidence remains limited.
Key Innovation: Soil Dynamics and Earthquake Engineering uses scaled shaking-table tests to evaluate cumulative damage under repeated seismic excitations.
47. CFSS strut system: A prefabricated method with low-carbon, economic and safety performance in deep excavation
Core Problem: Cast-in-place struts in deep excavations are slow, labor-intensive, and waste-generating.
Key Innovation: Transportation Geotechnics introduces a prefabricated concrete-filled steel strut system for deep excavation with structural, economic, and carbon-performance evaluation.
48. Failure mechanisms of tunnel models with structural planes under different connectivity ratios
Core Problem: Tunnel stability is strongly controlled by the connectivity of structural planes, but interaction mechanisms are underexplored.
Key Innovation: Transportation Geotechnics tests tunnel models with varying structural-plane connectivity to characterize failure mechanisms.
49. Disturbance-driven reorganization of microbial and physical controls on soil organic carbon in karst soils
Core Problem: Karst soils have shallow profiles and strong hydrological connectivity, making soil carbon vulnerable to disturbance.
Key Innovation: Catena analyzes how fire and cultivation reorganize microbial and physical controls on soil organic carbon in karst systems.
50. HYGRID-M: A grid-based hydrological balance model for water management at River Basin District scale
Core Problem: Basin-scale water management needs spatially explicit water-balance models that can represent heterogeneous soils and land use.
Key Innovation: Catena introduces HYGRID-M, a grid-based monthly hydrological balance model for River Basin District-scale water management.
51. Depth-dependent controls on soil saturated hydraulic conductivity in the Loess Plateau
Core Problem: Depth-dependent saturated hydraulic conductivity below 1 m remains poorly characterized in Loess Plateau catchments.
Key Innovation: Journal of Hydrology quantifies 0-500 cm vertical hydraulic-conductivity profiles and landscape controls in a small Loess Plateau watershed.
52. Physical ventilation failure drives rapid reservoir hypoxia during heatwaves
Core Problem: Rapid reservoir hypoxia during heatwaves is hard to attribute to biological demand versus physical ventilation failure.
Key Innovation: Journal of Hydrology combines spatial surveys and high-frequency time series to show that oxygen-supply reduction dominates rapid hypoxia in the Three Gorges Reservoir system.
53. Capturing temporal shifts in forest soil-atmosphere feedbacks: evidence from a Mediterranean catchment
Core Problem: Mediterranean catchments face water scarcity, but lead-lag dynamics among soil moisture, precipitation, vapor pressure deficit, and sap flow remain unclear.
Key Innovation: Journal of Hydrology uses wavelet analysis to capture temporal shifts in forest soil-atmosphere feedbacks.
54. Pixel-Level Uncertainty Quantification for Land Surface Temperature Retrieved from MODIS Thermal Infrared Data (2003–2023)
Core Problem: Long-term land-surface-temperature records need pixel-level retrieval uncertainty for climate, hydrology, and ecosystem modelling.
Key Innovation: Remote Sensing develops uncertainty quantification for MODIS thermal-infrared land-surface temperature from 2003-2023.
55. Enhancing Directional Wave Spectra Retrieval from Sentinel-1A SAR Wave Mode Under Strong Cut-Off Distortions via Prior-Knowledge-Integrated Machine Learning
Core Problem: Sentinel-1 ocean-wave retrieval suffers under strong azimuth cut-off distortions and unresolved wind-sea components.
Key Innovation: Remote Sensing integrates prior knowledge with machine learning to reconstruct directional wave spectra from Sentinel-1 SAR wave mode.
56. Cross-Modality Spectral Expansion Combined with Physical–Semantic Dual Priors for Cloud Detection in GF-1 Imagery
Core Problem: Cloud detection in GF-1 imagery is difficult because missing SWIR bands cause cloud-snow confusion.
Key Innovation: Remote Sensing combines cross-modality spectral expansion with physical and semantic priors to improve cloud detection.
57. Temporal Variations in $L$-Band Reflectivity Profiles of a Boreal Forest During Growth and Drought
Core Problem: Links between tree water stress, transpiration, and L-band reflectivity remain underused for forest drought monitoring.
Key Innovation: IEEE JSTARS models time-varying L-band reflectivity and attenuation in a boreal forest during growth and drought.
58. WGAST: Weakly Supervised Generative Network for Daily 10 m Land Surface Temperature Estimation via Spatio-Temporal Fusion
Core Problem: Daily 10 m land-surface-temperature estimation is limited by the tradeoff between spatial and temporal resolution.
Key Innovation: IEEE JSTARS proposes a weakly supervised generative spatiotemporal fusion network for daily 10 m LST estimation.
59. CAAMFormer: Cross-Attention Augmented Mamba Network for Remote Sensing Semantic Segmentation
Core Problem: High-resolution satellite scenes contain complex backgrounds and scale variation that challenge semantic segmentation.
Key Innovation: IEEE JSTARS introduces CAAMFormer, a cross-attention augmented Mamba network for remote-sensing semantic segmentation.
60. Robust detector-free multimodal image matching based on visual model guidance and gated attention
Core Problem: Multimodal image matching is vulnerable to geometric distortion and nonlinear radiometric differences.
Key Innovation: ISPRS Journal of Photogrammetry and Remote Sensing uses visual-model guidance and gated attention for detector-free multimodal image matching.
61. Ocean color remote sensing: From 2D legacy to the 3D, AI-driven future
Core Problem: Ocean-color remote sensing is moving beyond 2D legacy products toward richer AI-enabled three-dimensional observation.
Key Innovation: ISPRS Journal of Photogrammetry and Remote Sensing reviews the evolution of ocean-color remote sensing and its transition toward 3D, AI-driven monitoring.
62. Study on real-time forecast of tunnel fuel and electric vehicle fire states based on multimodal fusion learning
Core Problem: Tunnel fire-state perception fails when heat sensors are damaged and video is obstructed by smoke.
Key Innovation: Tunnelling and Underground Space Technology proposes multimodal fusion learning to estimate tunnel fire heat-release rate in real time.
63. Improving generalizability of data driven-based fire temperature field reconstruction models in large spaces through physics-guided domain adaptation
Core Problem: Data-driven fire-temperature reconstruction models often fail across different geometries, fire locations, and ventilation conditions.
Key Innovation: Tunnelling and Underground Space Technology integrates physics-guided domain adaptation with LSTM residual learning for transferable fire-temperature-field reconstruction.
64. Comparative study on combustion characteristics of dual-fire sources in naturally ventilated tunnels: Effects of fuel type and spacing
Core Problem: Multi-source tunnel fires can behave nonlinearly when different fuels and spacings interact.
Key Innovation: Tunnelling and Underground Space Technology experimentally compares dual ethanol and lithium-ion-battery/ethanol fire-source configurations in naturally ventilated tunnels.
65. A Shapley-value cooperative game-based risk decision model for natural gas pipeline vulnerability
Core Problem: Pipeline retrofitting decisions must prioritize interacting vulnerabilities under limited budgets and outage windows.
Key Innovation: Reliability Engineering & System Safety combines TOPSIS with Shapley-value cooperative game analysis to produce explainable pipeline vulnerability rankings.
66. Effects of cumulative exposures to climate-related disasters on social capital
Core Problem: As climate disasters become consecutive and overlapping, their cumulative effects on social capital and recovery capacity remain uncertain.
Key Innovation: International Journal of Disaster Risk Reduction uses longitudinal Australian data to analyze how repeated climate-related disaster exposure changes social-capital access and capacity.