TerraMosaic Daily Digest: July 11, 2026
Daily Summary
Failure studies now connect crack-scale constitutive physics to slope-scale precursors and long-lived landscape effects. Two rock-mechanics papers resolve compressive-shear fracture with a modified Mohr-Coulomb phase-field model and identify mixed-mode thresholds for unstable rock masses. Field evidence links hypermobile earthquake earthflows to irrigation-saturated farmland, shows anchor-force changes preceding acceleration on a rainfall-affected cut slope, rejects uniform weak-layer representations of basal rock-slope shear zones, and documents persistent post-debris-flow soil compaction and delayed vegetation recovery.
Hydrological risk is increasingly treated as a state- and network-dependent process. Time-lapse ground-penetrating radar resolves contrasting three-dimensional preferential-flow networks beneath broadleaf and coniferous hillslopes. State-conditioned residual learning reduces streamflow error most strongly in upper-tail flows, while post-deployment diagnostics expose coverage and redundancy in urban-drainage sensor networks. At larger scales, rural runoff, transport-network fragility, evacuation routes, reservoirs, and sediment pathways transmit local forcing through connected systems. REDES 2.0 places these processes against more than 2.4 million erosive storms and an approximately 9% increase in mean European rainfall erosivity.
Several studies make observational uncertainty a result rather than a footnote. Historical tropical-cyclone frequency trends can reverse sign with the chosen intensity threshold, and four coastal sea-surface-temperature products agree on only 12.34% of marine-heatwave days and 16.33% of events. Fourteen years of lidar observations constrain low-cloud trends, while circulation analysis separates daytime and nighttime mechanisms in the 2024 Yangtze-Huaihe heatwave. Climate-intervention experiments preserve snow controls more consistently than runoff responses. Across the modelling papers, water balance, chemistry, thermodynamics, fracture criteria, and network topology are imposed before prediction so that representation error is not mistaken for environmental change.
Key Trends
Five methodological shifts connect the physical, observational, and computational studies.
- Slope mechanics now span crack initiation to landscape recovery: Fracture criteria, mixed-mode thresholds, irrigation-conditioned liquefaction, anchor-force precursors, basal shear-zone structure, and post-failure soil recovery resolve distinct stages of instability.
- Hydrological prediction is becoming state- and network-aware: Preferential-flow connectivity, wet-dry residual separation, drainage-sensor influence, rural-urban runoff transfer, and transport-network fragility replace isolated-site assumptions.
- Trend claims are being tested against dataset disagreement: Intensity thresholds reverse historical cyclone trends, coastal temperature products identify different heatwave events, and active lidar constrains cloud changes that passive sensors can misclassify.
- Climate interventions expose regional thresholds: Cloud brightening and stratospheric aerosol experiments alter coupled climate and water responses unevenly; snow controls stabilize more consistently than runoff and teleconnection responses.
- Physical constraints are moving upstream in learned models: Water balance, chemistry, thermodynamics, fracture physics, dynamic boundaries, and sensor-network topology narrow the solution space before reconstruction or forecasting.
Selected Papers
The selected papers connect fracture mechanics, slope precursors, and post-failure recovery with state-dependent hydrology, networked flood risk, climate extremes, permafrost infrastructure, and explicit tests of observational uncertainty. This issue contains 209 selected papers from 2175 papers analyzed.
1. Analysis of anchor cable tension monitoring parameters and landslide instability mechanism induced by rainfall: a case study of the K3278 highway slope
Core Problem: Surface displacement can accelerate too late for operational warning, motivating precursors that respond earlier to rainfall-driven changes inside reinforced slopes.
Key Innovation: At the K3278 highway slope, anchor-cable tension anomalies lead displacement acceleration by seven and two days at monitored locations; coupled groundwater analysis also shows a nonlinear decline in factor of safety as the water table rises.
2. Hypermobility of seismic earthflows associated with modern intensive farming
Core Problem: The exceptional mobility of earthquake-triggered flows on gently sloping farmland is difficult to reconcile with conventional steep-slope landslide models.
Key Innovation: Comparison of the 2018 Palu and 2023 Jishishan events links irrigation-saturated soils, shaking-induced liquefaction, and extremely low H/L ratios of 0.011-0.027, identifying intensive farming as a shared precondition for seismic earthflows.
3. Current understanding of rock slope shear zones
Core Problem: Rock-slope models often treat basal shear zones as prescribed weak surfaces even though their origin, continuity, hydrology, and rock-bridge content are rarely known.
Key Innovation: The review synthesizes structural, geomorphic, geophysical, hydrogeological, and mechanical evidence into a workflow that moves from surface indicators to defensible subsurface shear-zone models.
4. Post-landslide changes in soil physical properties and vegetation recovery in a tropical mountain ecosystem of the Western Ghats, India
Core Problem: Tropical landslide corridors can remain physically degraded after visible greening, but coupled soil and vegetation recovery is rarely measured during the first years.
Key Innovation: Measurements along the Chooralmala debris-flow corridor show denser, rockier soils with lower water-holding capacity, a 31.9% severe vegetation decline, and only limited recovery after 1.5 years.
5. Circulation Depths of Meteoric Water in the Himalayas: Controls and Implications for Seismicity
Core Problem: The depth reached by meteoric water beneath the Himalayas remains uncertain, limiting tests of whether surface-derived fluids directly modulate regional seismicity.
Key Innovation: Hydrochemical and thermal constraints place most meteoric circulation at roughly 3-5 km, substantially shallower than the 5-25 km earthquake population and therefore separating near-surface fluid pathways from deeper seismic sources.
6. What Drives Precipitation Intensification During Zonal Atmospheric Rivers Landfalling in the Extratropical Andes?
Core Problem: Forecasting atmospheric-river rainfall in the extratropical Andes requires separating moisture supply from the dynamical and topographic controls that intensify precipitation at landfall.
Key Innovation: Analysis of 50 zonal atmospheric rivers identifies orographic ascent, atmospheric instability, and near-surface relative humidity as the principal controls on event-scale precipitation intensification.
7. Re-Examining Historical Trends of Tropical Cyclone Frequency
Core Problem: Historical reanalyses extend tropical-cyclone records beyond the satellite era, but sparse pressure observations and tracking choices can distort inferred frequency trends.
Key Innovation: Tests across two 20th Century Reanalysis versions, multiple trackers, and intensity thresholds show that even the trend sign can reverse and that large-scale environmental changes disagree between reanalysis products.
8. Significant Rural Contributions to Severe Coastal Urban Flooding Through Flow Connectivity
Core Problem: Coastal-city flood models commonly emphasize runoff generated inside the urban boundary and underrepresent connected rural source areas.
Key Innovation: Hydrodynamic attribution shows that rural runoff can dominate severe urban inundation once depths exceed about 1.7 m, while wetlands absorb more than 90% of the compounded contribution in the tested system.
9. Enhancing long-term reservoir inflow forecasting: an integrated approach combining switch prediction method, ensemble rainfall forecasts, and machine learning techniques
Core Problem: Typhoon reservoir forecasts inherit large uncertainty from rainfall ensembles and must remain stable across a 72-hour lead time.
Key Innovation: A switch-prediction method dynamically selects meteorological inputs before LSTM inflow forecasting; for Typhoon Soudelor it reaches a 178.8 cubic-metre-per-second MAE and 0.87 efficiency at the observed peak event.
10. Influential factors of soil water content in cut slope during ecological restoration: An experimental study
Core Problem: Ecological cut slopes must retain water for vegetation without allowing meteorology, topography, or restoration treatments to compromise structural stability.
Key Innovation: Sixteen instrumented model slopes separate seasonal and event-scale soil-water controls and quantify how restoration mode modifies moisture distribution under the same weather forcing.
11. Blind-prediction of the dynamic behavior of 3D-printed masonry-like structures under shake-table loading via a high-fidelity numerical modeling approach
Core Problem: Predicting the nonlinear shaking response of masonry-like structures without calibrating to the target test remains a stringent validation problem.
Key Innovation: A blind high-fidelity numerical prediction is tested against shake-table experiments on 3D-printed masonry analogues, directly evaluating whether the modelling chain transfers from material characterization to structural dynamics.
12. Resilience assessment of transportation systems subjected to hurricane-induced storm surge
Core Problem: Road-network resilience to hurricanes depends on bridge fragility, traffic redistribution, repair cost, and recovery time, which are often assessed separately.
Key Innovation: A storm-surge framework couples regional bridge fragilities with full traffic simulations in two Texas coastal testbeds and evaluates mitigation through a multidimensional resilience index.
13. Wildfire impact potential along routes shapes mobility patterns during the evacuation period
Core Problem: Wildfire evacuation studies usually characterize destination risk without measuring the changing hazard encountered along the route.
Key Innovation: Anonymized mobility and time-resolved fire data from the 2023 McDougall Creek fire show that fire weather and structural vulnerability along routes explain trip patterns better than destination conditions.
14. An expert-informed, open-data-driven framework for profiling multisystem co-disruptions in critical urban systems in Attica
Core Problem: Comparing disruptions across urban infrastructure systems is difficult when open datasets differ in timing, admissibility, baseline, and proxy quality.
Key Innovation: A six-gate source protocol and integrated fragility-impact synthesis produce auditable co-disruption signatures for four hydrometeorological event windows across five systems in Attica.
15. A framework for assessing dam failure consequences in cascade reservoir systems under the domino effect
Core Problem: Failure consequences in cascade reservoirs cannot be estimated by treating each dam independently because upstream breaches can trigger downstream domino effects.
Key Innovation: The framework propagates breach waves through the reservoir chain and attributes exposure and consequence across successive failures, enabling system-level rather than single-dam risk assessment.
16. High-precision extraction and dynamics of thermokarst lakes in the Three Rivers Source Region using the SVM coupled edge-OTSU based on Sentinel-1 data
Core Problem: Optical mapping of thermokarst lakes confuses snow-covered depressions with water, biasing inventories in permafrost terrain.
Key Innovation: A Sentinel-1 SVM plus edge-OTSU workflow reaches 95.87% overall accuracy and maps 31,800 lakes in the Three Rivers Source Region in 2024, revealing expansion from 2015 to 2024.
17. REDES 2.0: A pan-European rainfall erosivity dataset showing intensifying erosivity trends
Core Problem: European rainfall-erosivity estimates have lacked a harmonized event database dense enough to resolve regional trends and extremes.
Key Innovation: REDES 2.0 compiles more than 2.4 million erosive events from 9,138 stations; the resulting pan-European record shows an approximately 9% mean R-factor increase and positive trends at 65% of stations.
18. Numerical modelling of strain-localisation in compressed snow samples
Core Problem: Compaction-band localization in porous snow is a precursor to slab failure but is difficult to reproduce with rate-independent constitutive models.
Key Innovation: An elasto-visco-plastic snow model with sintering and bond degradation reproduces confined-compression band formation and maps the strain-rate conditions that promote avalanche-relevant localization.
19. Design, optimization and field application of a pressurized absorption refrigeration system for heat recovery and thermal hazard mitigation in high geothermal tunnels
Core Problem: High-geothermal tunnels expose workers to severe heat while continuously discharging hot water whose energy is usually wasted.
Key Innovation: A pressurized absorption-refrigeration system uses tunnel inflow water for local cooling, is thermodynamically optimized, and is demonstrated in a long railway tunnel.
20. Unraveling the rainfall-waterlogging propagation lag in urban systems: mechanisms, response patterns, and source tracking evidence
Core Problem: Urban waterlogging forecasts often omit the delay and source pathways between rainfall input and inundation response.
Key Innovation: Event analysis separates response patterns, quantifies rainfall-waterlogging lag, and traces source contributions, converting a simple rainfall threshold into a process-aware warning relation.
21. Three-dimensional site response and dynamic stress concentration around a spherical cavity in an unsaturated porous half-space under SV-wave incidence
Core Problem: Subsurface cavities concentrate seismic stress, but three-dimensional response in partially saturated porous ground is poorly represented by simplified elastic models.
Key Innovation: An analytical 3D solution resolves SV-wave scattering, pore-fluid coupling, and dynamic stress concentration around a spherical cavity in an unsaturated half-space.
22. Analytical solution for the longitudinal seismic response of multi-layer lined tunnels considering shear interaction at ground-lining and layered-lining interfaces
Core Problem: Layered tunnel linings can slip or shear at both ground-lining and lining-lining interfaces during longitudinal seismic deformation.
Key Innovation: The analytical solution represents both interfaces explicitly, showing how interaction stiffness redistributes axial and shear demand through multilayer linings.
23. Study on seismic displacement response test and deformation prediction model of composite panel reinforced soil retaining wall
Core Problem: Seismic deformation of reinforced-soil retaining walls depends on coupled panel and backfill movement that static displacement formulas do not capture.
Key Innovation: Shaking tests support a deformation-prediction model for composite-panel reinforced-soil walls and link observed displacement modes to excitation characteristics.
24. A representative seismic velocity spectrum without constant-velocity assumption
Core Problem: Constant-velocity assumptions flatten the frequency dependence of earthquake ground motion and can bias spectrum-compatible record representation.
Key Innovation: A representative seismic velocity spectrum retains frequency-dependent velocity content, providing an alternative basis for ground-motion characterization.
25. Cleaning Air Pollution Over Central Europe Enhances Extreme Precipitation
Core Problem: The response of sub-daily precipitation extremes to aerosol cleanup is poorly constrained because aerosol and greenhouse-gas effects are difficult to separate in observations.
Key Innovation: Clean-air simulations over Central Europe increase the 99th, 99.9th, and 99.97th percentiles of hourly precipitation by 2.5%, 6.4%, and 7.9%, respectively, revealing a disproportionate response in the most intense events.
26. Low Cross-Product Agreement in Global Coastal Marine Heatwaves (1982-2023) and Implications for Trend Attribution
Core Problem: Coastal marine-heatwave records derived from different sea-surface-temperature products can identify different days, events, and drivers of change.
Key Innovation: Four daily products are compared at 9,946 coastal locations; all-product agreement averages only 12.34% for days and 16.33% for events, while trend attribution diverges more strongly than the trends themselves.
27. Precipitation Intensity Increased Across the South Pacific Convergence Zone in the Industrial Age
Core Problem: Coupled models disagree on tropical Pacific rainfall because the South Pacific Convergence Zone is difficult to represent and instrumental records are short.
Key Innovation: Biomarker-isotope reconstructions from five island lakes extend precipitation history by a millennium and identify an industrial-era intensification beginning around 1820.
28. Ecohydrologic Processes Modify Urban Rainfall Intensification in Land-Atmosphere Simulations
Core Problem: Urban land-atmosphere models omit fine-scale canopy, pavement, and runoff-routing processes that can modify convective rainfall.
Key Innovation: Milwaukee ensemble simulations show that explicit urban ecohydrology produces cooler, wetter pre-storm conditions and improves rainfall agreement in a known intensification zone.
29. The Role of the Heterogeneous Variation of East Asian Subtropical Jet in the 2024 Heatwave Over Yangtze-Huaihe River Basin
Core Problem: The circulation mechanisms separating daytime, nighttime, and compound heatwaves over the Yangtze-Huaihe River Basin remain difficult to isolate.
Key Innovation: Three 2024 heatwave periods show that northward jet displacement favors daytime warming, whereas cyclonic jet meanders suppress nighttime cooling through convection and enhanced downward radiation north of the basin.
30. Characterizing Watershed Responses and Reservoir Operational Flexibilities: Analyses to Support Forecast-Informed Reservoir Operations (FIRO) Planning and Assessments
Core Problem: Forecast-informed reservoir operation requires enough lead time and storage flexibility to act on inflow forecasts without increasing downstream flood risk.
Key Innovation: A transferable diagnostic links precipitation phase, soil moisture, runoff generation, inflow timing, available storage, and safe release capacity to screen reservoirs for FIRO feasibility.
31. Omitting Frozen-Soil Effects in Hydrological Models Introduces Biases in Estimated Runoff Sensitivity to Climate Change in Cold Basins
Core Problem: Cold-basin runoff projections can appear accurate historically while remaining process-biased if frozen-soil controls on subsurface partitioning are omitted.
Key Innovation: Paired tracer-aided models show that omission lowers simulated subsurface contribution from 68.5% to 41.3% and more than doubles temperature-driven runoff sensitivity.
32. An Interpretable Deep Learning Framework for Evaluating Sensor Networks in Urban Drainage Systems
Core Problem: Existing urban-drainage sensor studies emphasize pre-installation placement and provide limited diagnosis of coverage, redundancy, and influence after deployment.
Key Innovation: Sparse water-level reconstruction and learned directional dependencies support three interpretable network indicators; mean Nash-Sutcliffe efficiency exceeds 0.85 and inferred influence rankings correlate with perturbation tests at 0.60-0.75.
33. Rainfall Field Reconstruction From Commercial Microwave Links via a Conditional Diffusion Model
Core Problem: Commercial microwave links sample rainfall densely along paths, but converting their attenuations into spatially coherent rainfall fields is an underdetermined inverse problem.
Key Innovation: A conditional diffusion model reconstructs rainfall fields from link observations while representing multiple plausible spatial solutions rather than returning a single deterministic interpolation.
34. Large Scale Connectivity of River Networks During Rainfall Events
Core Problem: River-network connectivity during individual storms is rarely quantified at basin scale, particularly where channels are intermittent.
Key Innovation: Topographic structural networks and runoff-correlation functional networks distinguish synchronous from sequential connectivity and relate storm classes to dominant flow pathways in Walnut Gulch.
35. From restoration to defence: Integrating artificial strategies and salt marshes to construct resilient ecosystem-based coastal defences
Core Problem: Salt-marsh restoration alone is difficult on exposed muddy coasts, where waves and elevated water levels can prevent vegetation establishment before ecological protection develops.
Key Innovation: The study combines an artificial shelter with salt-marsh vegetation and tests decadal elevated-water and enhanced-wave scenarios, maintaining more than 70% wave attenuation in representative sheltered-zone cases.
36. Differential shoreline dynamics surrounding a sedimentary headland marsh system in a microtidal deltaic coast
Core Problem: Sediment-starved delta marshes depend on local resuspension, but neighbouring shorelines can erode and accrete differently under the same wind climate.
Key Innovation: Field hydrodynamics, topographic surveys, vegetation measurements, and Sentinel-2 imagery trace wind-driven sediment delivery and contrasting shoreline response around a Terrebonne Bay headland.
37. A physically guided deep learning reconstruction of terrestrial water storage anomalies at 0.1° across China
Core Problem: Terrestrial water-storage records are too coarse or incomplete for resolving regional drought and groundwater dynamics at decision-relevant scales.
Key Innovation: A physically guided deep-learning reconstruction combines water-balance constraints with satellite observations to produce 0.1-degree terrestrial water-storage anomalies across China.
38. EXSoDOS 1.0: downscaling of weather extremes shifts for ensemble climate projections using ground-based measurements, reanalysis and stochastic modelling
Core Problem: Regional risk assessments need local changes in weather extremes, but global climate ensembles remain too coarse and conventional downscaling can distort tail behaviour.
Key Innovation: EXSoDOS combines station observations, reanalysis, and stochastic modelling to downscale ensemble shifts in extremes while retaining local distributional structure.
39. Pre-training for deep statistical climate downscaling: enhancing consistency and robustness across regional datasets
Core Problem: Deep statistical downscaling often loses consistency when transferred between regions with different observing density and climate regimes.
Key Innovation: Cross-regional pre-training improves robustness and spatial consistency before regional fine-tuning, reducing dependence on a single local training archive.
40. Volcanoes and wildfires contributed to increased stratospheric humidification
Core Problem: The recent rise in stratospheric water vapour cannot be explained by surface warming alone.
Key Innovation: Satellite observations and climate simulations attribute roughly one-third of the trend since 2005 to aerosol-mediated effects of moderate volcanic eruptions and extreme wildfires.
41. Targeted marine cloud brightening weakens subsequent El Niño
Core Problem: Regional marine cloud brightening could alter ENSO, but its effectiveness and delayed teleconnections have not been tested against a natural analogue.
Key Innovation: Simulations reproduce the Australian-wildfire cloud-brightening response and show that only early, sustained intervention substantially weakens El Nino, sometimes advancing a subsequent La Nina.
42. A chemistry-informed deep learning network for mitigating the stratospheric OH data gap
Core Problem: Global stratospheric hydroxyl observations have had a major satellite gap since 2009, limiting diagnosis of ozone chemistry and exceptional injections.
Key Innovation: DRCAT embeds chemical relationships in a deep network to reconstruct continuous OH profiles and generalizes to the anomalous enhancement after the 2022 Hunga eruption despite training on only two years.
43. Flood Regulation Service Responses to Urban Green Space Change in a Plateau Valley City: A Case Study of Lhasa, China
Core Problem: Rapid urban growth in plateau valleys changes runoff pathways, but the flood-regulation value of green space is rarely quantified through time.
Key Innovation: Multi-temporal remote sensing and runoff modelling map how changes in Lhasa's urban green space redistribute flood-regulation services across the valley city.
44. RadarEchoMamba: A Fast, High-Fidelity Pyramidal Bidirectional Mamba Model for Radar Echo Extrapolation
Core Problem: Radar nowcasting models blur fine precipitation structures when temporal information decays or global attention becomes computationally expensive.
Key Innovation: RadarEchoMamba combines a pyramidal encoder, bidirectional spatiotemporal state-space modelling, and input-driven priors to preserve fine echo structure with lower latency than strong Transformer baselines.
45. Effects of different tillage types on soil erosion in the hilly red soil region of Southeastern China
Core Problem: Soil-erosion response to tillage on red-soil hillslopes is poorly constrained during the high-intensity storms that dominate annual loss.
Key Innovation: Twelve monitored storms across five plot treatments show that bare soil produces orders-of-magnitude more sediment and soil loss than managed slope configurations.
46. A Phase Field Model for Rock Fracture Based on the Modified Mohr-Coulomb Failure Criterion
Core Problem: Conventional phase-field models struggle to reproduce the tension-compression asymmetry and compressive-shear fracture patterns of rock.
Key Innovation: A stress-based energy split embeds a modified Mohr-Coulomb criterion, separate tensile and compressive thresholds, and a strength-dependent shear driving force, with validation extending from benchmark specimens to tunnel failure and rock-slope instability.
47. Computational Fracture Mechanics Modeling and Multivariate Analysis of Mixed-Mode I-II Crack Propagation in Unstable Rock Masses
Core Problem: Fractured rock-slope assessment needs a constitutive description that separates crack geometry from open, closed, and frictional interface behavior.
Key Innovation: A two-parameter mixed-mode framework identifies a shear-dominated regime below a K-IIC/K-IC ratio of 1.35 and a tensile-to-shear transition at crack inclinations of 55-60 degrees.
48. Experimental and mechanical behavior investigation on self-centering braces with multi-stage yield energy dissipation
Core Problem: Seismic braces must dissipate energy without leaving large residual drift, yet conventional self-centering devices often provide only one yielding stage.
Key Innovation: A disc-spring self-centering brace with staged metallic yielding is validated under low-cycle loading, linking pre-compression, spring stiffness, and plate width to recentering and energy dissipation.
49. Examining the short-term impact of evacuation orders on traffic conditions: Evidence from Hurricane Ian
Core Problem: Evacuation orders can save lives while abruptly overloading roads, yet their immediate local traffic effects are difficult to isolate from the evolving hurricane itself.
Key Innovation: High-resolution mobility data and a regression-discontinuity design quantify changes in speed and travel-time reliability after Hurricane Ian evacuation orders in Lee County.
50. Multi-criteria decision-making for enhancing seismic resilience of buildings via optimization of isolation system and nonstructural components
Core Problem: Life-safety design does not ensure that isolated buildings and their nonstructural systems remain functional after earthquakes across multiple intensity levels.
Key Innovation: A progressive resilience framework uses multi-objective optimization for isolation parameters, then multi-criteria selection to reconcile response reduction, bearing feasibility, and nonstructural-component strengthening.
51. Detection of oriented lunar rockfalls and kinematic analysis in Tycho Crater
Core Problem: Lunar rockfall mapping is hindered by small, elongated deposits with arbitrary orientation and by the absence of dense manual inventories.
Key Innovation: An orientation-aware coarse-to-fine detector reaches 63.0% AP50 and 78.9% recall in Tycho Crater, then uses mapped trajectories for kinematic analysis of source slopes and runout.
52. From badland to bushland? Analysis of geomorphic process dynamics and vegetation development in a sub-humid calanchi area based on high-resolution UAS data (2014-2024)
Core Problem: Badland recovery is governed by episodic erosion and vegetation establishment, but decadal measurements rarely resolve their coupled thresholds.
Key Innovation: Repeated high-resolution UAS surveys from 2014 to 2024 reveal declining affected area but persistent event-scale erosion, vegetation colonization of depositional zones, and infiltration-driven gravitational bulging.
53. Multifractal signatures of microtopography under consecutive rainfall: Contrasting rock fragment cover vs. embedded configurations on karst slopes
Core Problem: Rock-fragment configuration alters runoff and sediment production on karst slopes, yet mean roughness alone misses multiscale surface reorganization.
Key Innovation: Sequential rainfall experiments and photogrammetric DEMs show configuration-dependent sediment response and multifractal changes, including substantially larger heterogeneity growth under surface cover.
54. A novel geochemical enhanced measurement while drilling method for adverse geology identification: GE-MWD
Core Problem: Conventional measurement-while-drilling signals can confuse operational variation with adverse geological changes ahead of excavation.
Key Innovation: GE-MWD adds geochemical evidence to drilling mechanics so weak, fractured, or water-bearing ground can be identified with a more diagnostic multimodal signature.
55. Divergent 3D dynamics of subsurface lateral preferential flow network in broadleaf versus coniferous hillslopes: insights from time-lapse GPR
Core Problem: Subsurface lateral preferential flow is difficult to observe in three dimensions, obscuring how forest type changes hillslope connectivity and runoff generation.
Key Innovation: Time-lapse ground-penetrating radar and controlled infiltration map contrasting flow networks: broadleaf hillslopes activate rapidly and deeply, whereas coniferous hillslopes respond later, more shallowly, and more incrementally.
56. Toward understanding tidal flat regime shifts: Contrasting hydro-sediment dynamics revealed by paired observations of two tidal flats
Core Problem: Coastal tidal flats can switch abruptly from accretion to erosion, yet sediment concentration alone does not diagnose which regime is active.
Key Innovation: Paired high-frequency observations show that wave stress drives persistent export at the eroding site, whereas morphology creates early-flood deposition windows at the accreting site.
57. Short-term changes in slope vegetation patch patterns and their regulation on sediment connectivity in an arid area
Core Problem: Vegetation cover alone does not explain how patch geometry interrupts runoff and sediment pathways on arid slopes.
Key Innovation: Three years of UAV, terrestrial-laser-scanning, and field observations show that uniform patches reduce runoff by 40.46% and sediment yield by 53.48%, while fragmentation increases sediment connectivity.
58. Seasonal controls, recharge thresholds, and hydrological memory in karst groundwater dynamics revealed by interpretable deep learning
Core Problem: Karst recharge is nonlinear and seasonally dependent, but prediction models rarely expose thresholds or memory in groundwater response.
Key Innovation: An interpretable LSTM reproduces Baotu Spring levels with 0.93 Nash-Sutcliffe efficiency and resolves seasonal attribution, rainfall-response ranges, and winter memory extending to about five months.
59. SG-PINN: a spatial-gating physics-informed neural network with fourier features to mitigate spectral bias and nonphysical far-field oscillations in geotechnics
Core Problem: Geotechnical PINNs can fit governing equations yet develop spectral bias and nonphysical oscillations away from the constrained region.
Key Innovation: SG-PINN combines Fourier features with spatial gating to represent sharp local behaviour while suppressing spurious far-field responses.
60. Thermo-hydro-mechanical modeling of single-phase fluid flow in heterogeneous porous media with micro-fractures using physics-informed neural network
Core Problem: Coupled heat, fluid, and deformation in fractured porous media is costly to resolve when material properties vary sharply at micro-fractures.
Key Innovation: A physics-informed network embeds thermo-hydro-mechanical equations and heterogeneous fracture properties into one differentiable solution framework.
61. Modelling hydro-mechanical coupling in geotechnical problems using the material point method with dynamic boundary conditions
Core Problem: Large-deformation geotechnical simulations need hydraulic boundary conditions that remain valid as the material domain moves.
Key Innovation: A material-point formulation introduces dynamic hydro-mechanical boundaries that follow the evolving geometry and preserve coupled mass and momentum transfer.
62. Implementation and validation of a reproducible coupled hydro-mechanical CEL framework for large-deformation modelling of saturated soils
Core Problem: Coupled Eulerian-Lagrangian models of saturated-soil deformation are difficult to reproduce because coupling and validation choices vary between implementations.
Key Innovation: The study documents and validates a reproducible hydro-mechanical CEL workflow for large deformation, making numerical assumptions and benchmark behaviour explicit.
63. Seismic performance evaluation of bridge isolated by damping enhanced-friction pendulum system considering variable friction effect
Core Problem: Bridge-isolation predictions are sensitive to friction that changes with pressure and velocity, but fixed-friction models suppress that variability.
Key Innovation: Four damping-enhanced friction-pendulum specimens establish variable-friction behaviour, which is embedded in a validated OpenSees bridge model to quantify isolation displacement and pier-force trade-offs.
64. Mechanism and parametric analysis on cooperative seismic mitigation of base-isolated structures equipped with negative stiffness amplifying damper based on a distributed parameter model
Core Problem: Base isolation can reduce superstructure demand while leaving excessive isolator motion, especially under long-period pulse-like near-fault shaking.
Key Innovation: A distributed-parameter model couples the isolated structure to a tuned negative-stiffness amplifying damper and maps the parameter ranges that provide cooperative force and displacement control.
65. Out-of-plane seismic behavior of non-framed unreinforced masonry walls with pre-existing seismic damage: numerical study via a block-based modeling approach
Core Problem: Unreinforced masonry walls may face out-of-plane shaking after accumulating in-plane earthquake damage, but this interaction is rarely represented for non-framed buildings.
Key Innovation: A block-based numerical model carries explicit in-plane pre-damage into subsequent out-of-plane loading and resolves how crack state changes residual capacity and collapse response.
66. Performance analysis of an SMA negative-stiffness double-concave friction pendulum bearing in reinforced concrete frame structure
Core Problem: Friction-pendulum bearings struggle to limit both peak isolation displacement and residual drift during strong earthquakes.
Key Innovation: Horizontal shape-memory-alloy cables add superelastic recentering to a negative-stiffness double-concave bearing, producing an adaptive energy-dissipation mechanism evaluated in a reinforced-concrete frame.
67. Experimental study on the seismic response of square precast RC columns with grouted sleeves strengthened by an ECC jacket and its design-oriented bearing-capacity prediction method
Core Problem: Grouted-sleeve connections can localize damage in precast reinforced-concrete columns, limiting ductility under cyclic earthquake loading.
Key Innovation: Cyclic tests evaluate an engineered-cementitious-composite jacket around square sleeve-connected columns and support a design-oriented bearing-capacity model for the strengthened section.
68. An efficient finite element framework for nonlinear seismic soil-foundation-structure interaction in shallow-founded squat masonry structures
Core Problem: Fully nonlinear 3D simulation of squat masonry buildings with soil-foundation-structure interaction remains expensive when masonry, interfaces, and soil all yield.
Key Innovation: An OpenSees framework integrates nonlinear masonry, contact interfaces, soil plasticity, coupling, and absorbing boundaries in one direct model, then benchmarks its accuracy and computational demand.
69. Design procedure for viscoelastic energy dissipation devices in seismic strengthening of unreinforced masonry buildings
Core Problem: Stiffness-based strengthening of historic unreinforced masonry can conflict with conservation requirements while leaving facades vulnerable to out-of-plane failure.
Key Innovation: A simplified design procedure sizes geometric-nonlinear viscoelastic devices to dissipate facade motion with limited intervention, connecting device geometry to target seismic demand.
70. Tall building isolated by segmented friction pendulum bearing: dynamic response analysis and multi-objective optimization
Core Problem: Continuously varying curvature and friction are difficult to manufacture and control in variable-frequency pendulum isolators.
Key Innovation: A two-disc segmented friction-pendulum bearing replaces continuous surfaces with independently designed segments and uses dynamic analysis plus multi-objective optimization to select practical configurations for tall buildings.
71. Seismic responses of an FPB-isolated railway continuous beam bridge under pulse-type motions: Shaking table tests and simulations
Core Problem: Pulse-type ground motions can drive large displacement and velocity-sensitive friction in isolated railway bridges, outside the assumptions of constant-friction design models.
Key Innovation: Shake-table tests and simulations calibrate a velocity- and pressure-dependent friction law for pendulum bearings and trace its effect on bridge demand under pulse-like records.
72. Building-scale seismic performance and resonance mitigation of variable-curvature sliding isolators under pulse-type ground motions
Core Problem: Near-fault velocity pulses can excite resonance-like response and excessive bearing travel in base-isolated buildings.
Key Innovation: Building-scale tests and analysis evaluate variable-curvature sliding isolators, identifying curvature profiles that suppress resonant amplification while controlling isolation displacement.
73. Investigation of internal force redistribution in shear wall structures subjected to seismic action
Core Problem: Combining shear-wall types with different stiffness and deformation modes can redistribute seismic forces in ways that component-by-component design misses.
Key Innovation: The study quantifies force transfer among wall configurations under earthquake loading and identifies how layout governs local demand rather than assuming proportional sharing.
74. Three-dimensional continuum-based analytical model for longitudinal seismic response of shafts subjected to vertically incident P-waves
Core Problem: Longitudinal earthquake demand on hollow shafts depends on coupled radial and tangential soil contact that one-dimensional beam models cannot represent.
Key Innovation: A 3D continuum solution with independent normal and tangential interface elements spans weak contact to perfect bonding and maps frequency-dependent displacement, strain, stress, and energy dissipation.
75. Impact of plan layouts considering soil-structure interaction on seismic response of irregular adjacent steel buildings
Core Problem: Adjacent irregular steel buildings are vulnerable to seismic pounding when asymmetric plans and flexible soil alter their relative motion.
Key Innovation: Nonlinear gap elements and an equivalent soil-structure interaction model compare plan layouts, showing how configuration changes global response, separation demand, and pounding potential.
76. Adjusted design and experimental feasibility of an enhanced damping-isolated tuned liquid damper considering soil-structure interaction
Core Problem: Tuned liquid dampers use only part of their liquid mass efficiently and lose effectiveness under broadband excitation, particularly when soil flexibility shifts structural frequencies.
Key Innovation: An enhanced damping-isolated liquid damper is optimized with a coupled acoustic-structure surrogate, experimentally checked, and then tested across soil-structure interaction conditions.
77. Wood in Patagonian headwater streams: A window into the colonial and wildfire caused legacy of large wood in temperate stream ecosystems
Core Problem: Wildfire and forest clearing leave long-lived wood legacies in headwater channels, but their geomorphic expression differs between dry deciduous and humid evergreen forests.
Key Innovation: Paired Patagonian watersheds show consistently greater in-stream large-wood storage at impacted sites and connect present channel structure to 50-70 years of fire history.
78. Monthly runoff and sediment fluxes in a typical arid and semiarid river basin: Trends, driving factors, and attribution analysis
Core Problem: Runoff and sediment flux in arid basins vary across several overlapping timescales, making annual trend attribution alone misleading.
Key Innovation: Signal decomposition and wavelet coherence separate monthly oscillations, phase lags, and the respective roles of precipitation, runoff, and human activity in the Ulan Mulun Basin.
79. High-Resolution Magnetic Signature of Lava Flows, Hydrothermal Field and Fossil Lava Lakes on the Equatorial East Pacific Rise
Core Problem: Young lava flows, drained lava lakes, collapse voids, and hydrothermal alteration are difficult to distinguish with conventional seafloor mapping.
Key Innovation: AUV magnetic and bathymetric surveys at 5 m altitude resolve three successive East Pacific Rise lava flows and alteration or void zones shallower than 10 m.
80. Burned Area Over Africa Reduced by Shortened Dry Season
Core Problem: African burned area is concentrated in the dry season, but the fire response to shifts in dry-season onset and duration has not been quantified consistently across the continent.
Key Innovation: A 1990-2023 climate record links delayed dry-season onset and increased early-season rainfall to shorter dry seasons and a marked 2003-2022 decline in burned area, strongest during peak fire months.
81. Trends and Sensitivities of Low-Cloud Cover and Top Height From Spaceborne Lidar Observations
Core Problem: Passive satellite retrievals can confuse low clouds with the surface, limiting direct tests of changes in cloud cover and cloud-top height.
Key Innovation: Fourteen years of spaceborne lidar observations reveal anticorrelated spatial trends in low-cloud cover and top height and quantify their sensitivity to sea-surface temperature and inversion strength.
82. Integrating Socioeconomic Withdrawals Into a Deep Learning Framework for High-Resolution Groundwater Storage Prediction in the Yellow River Basin
Core Problem: GRACE resolves basin-scale groundwater storage but misses local depletion, while statistical downscaling can misattribute human withdrawals to climate variability.
Key Innovation: A groundwater-balance-constrained deep-learning framework incorporates socioeconomic withdrawals, separating climate- and human-dominated depletion while reconstructing high-resolution storage change across the Yellow River Basin.
83. Hydrologic Sensitivity of Snow and Streamflow Dynamics to Climate Forcing With and Without Stratospheric Aerosol Intervention
Core Problem: Warming shifts snowpack and warm-season runoff from precipitation control toward temperature control, but the regional hydrological response to climate intervention is unresolved.
Key Innovation: Moving-window diagnostics in two Earth-system models show that stratospheric aerosol intervention largely stabilizes snow-control transitions, especially where cold-season temperatures lie between -4 and 0 C, while runoff responses remain heterogeneous.
84. Can State-Dependant Residual Modeling Improve Legacy Hydrological Model Simulations?
Core Problem: A single residual-correction model treats wet- and dry-state hydrological errors alike even though their sources and behavior differ.
Key Innovation: A two-stage, state-dependent deep-learning correction separates dry- and wet-period residuals, reducing RMSE by 59-91% overall and 65-86% for upper-tail flows in the Warren River catchment.
85. Fall Soil Moisture Modulates Snow-Streamflow Dynamics in the Colorado River Basin
Core Problem: Snowpack alone no longer explains Colorado River runoff reliably during prolonged drought, leaving the role of antecedent basin storage unresolved.
Key Innovation: Controlled VIC experiments attribute 69-77% of Upper Basin streamflow variability to 1 October soil moisture and identify spring precipitation as the dominant Lower Basin control.
86. Improving Daily Runoff Prediction Using a Novel Two-Step Post-Processing Method of Frequency Distribution Curve Correction
Core Problem: Hydrological forecasts contain both systematic offsets and distributional errors that single post-processing methods cannot correct simultaneously.
Key Innovation: A two-step statistical plus LSTM correction improves high-, medium-, and low-flow prediction across eight watersheds while reducing distributional divergence.
87. Does Antecedent Catchment Wetness Explain the Timing of Rainfall-Runoff Relationship Shifts?
Core Problem: Many catchments produce persistently less runoff after drought, but the climatic state governing entry into and recovery from these low-runoff regimes is unclear.
Key Innovation: A hidden-state analysis of 158 Australian catchments shows that antecedent wetness, especially in autumn, explains much of the timing and hysteresis of rainfall-runoff regime shifts.
88. Correlations versus machine learning models for prediction of Vs from CPT in fine-grained soil: a practice-oriented approach
Core Problem: Shear-wave velocity estimates from cone penetration tests in fine-grained soils must remain accurate with incomplete inputs and across sites, conditions under which black-box gains may not transfer.
Key Innovation: A practice-oriented comparison shows when tree ensembles, support-vector regression, neural networks, and conventional correlations are reliable under different CPT input combinations and data completeness.
89. Immersion Ratio-Driven Degradation and Load Upshift of Belled Uplift Piles under Groundwater Effects: Model Tests, Field Benchmark, and Design Relations
Core Problem: Groundwater rise progressively degrades belled uplift-pile capacity, but design relations do not represent the full loading-path dependence.
Key Innovation: Scaled tests and a field benchmark use a dimensionless immersion ratio to explain nonlinear capacity loss and the accelerated degradation beyond half immersion.
90. FBG-based dynamic characterization of a scaled offshore wind turbine foundation model under impact and recorded seismic excitations
Core Problem: Dynamic monitoring of offshore foundations must distinguish connection damage from changes in boundary condition and excitation direction.
Key Innovation: Fiber-Bragg-grating and accelerometer measurements on a 1:30 model recover stable seismic frequencies and identify a double spectral peak after controlled bolt loosening.
91. A Topology-Driven Cross-Modal Fusion Network for Winter Wheat Parcel Extraction From Optical and SAR Imagery
Core Problem: Fine agricultural parcels are difficult to delineate when optical imagery is cloud-limited and SAR geometry differs sharply from optical boundaries.
Key Innovation: A topology-driven dual branch aligns optical and SAR structure before fusion, preserving parcel connectivity and boundaries during winter-wheat extraction.
92. A Dual-Aggregation and Semantic Prototype Guided Network for Railway Point Cloud Segmentation
Core Problem: Railway point-cloud segmentation is degraded by severe class imbalance, repeated local geometry, and long connected infrastructure with complex topology.
Key Innovation: DA-Net combines dual feature aggregation with semantic prototypes so rare railway components retain both local geometry and network-scale context.
93. CAMELS-FI: hydrometeorological time series and landscape properties for 320 catchments in Finland
Core Problem: Large-sample hydrological modelling in Finland has lacked a harmonized archive linking streamflow and meteorological time series to catchment attributes.
Key Innovation: CAMELS-FI releases standardized hydrometeorological records and landscape descriptors for 320 Finnish catchments, extending cross-catchment benchmarking into a cold-region setting.
94. A self-supervised precipitation forecast verification based on contrastive learning
Core Problem: Pointwise precipitation verification over-penalizes small spatial offsets, whereas neighbourhood methods depend on hand-selected scales and thresholds.
Key Innovation: CLPFV learns a self-supervised contrastive representation of precipitation fields and compares forecasts in feature space, reducing dependence on event-specific verification rules.
95. A unified scheme for modeling saturation and infiltration excess runoff
Core Problem: Saturation-excess and infiltration-excess runoff are commonly represented by separate theories despite transitioning within the same storm and catchment.
Key Innovation: A unified soil-storage-distribution scheme allows both mechanisms to change dynamically in space and time and is validated across 181 US catchments.
96. Characterizing low and high flow spells and their temporal transitions using baseflow estimates
Core Problem: Flood and drought spells are often evaluated separately, obscuring how quickly catchments transition between extremes before hydrological recovery.
Key Innovation: A baseflow-informed mixed-threshold framework detects high- and low-flow spells across 643 French catchments and resolves their duration, transition time, and regional operating context.
97. Quantifying the response of water and carbon balances to land cover and climate extremes across Germany
Core Problem: National carbon and water assessments rarely quantify how land cover and antecedent storage mediate responses to exceptional floods and droughts.
Key Innovation: WaSSI simulations across Germany map joint water-yield and carbon-sequestration responses and show that previous-year hydrological buffers moderate the impacts of major climate extremes.
98. Assessing deficiencies in remotely sensed actual evapotranspiration (AET): introducing AET signatures
Core Problem: Remote-sensing evapotranspiration products can match annual totals while missing variability that matters for diagnosing drought and land-atmosphere exchange.
Key Innovation: Eight multi-timescale evapotranspiration signatures expose product-specific errors against 17 Australian flux towers that aggregate annual comparisons conceal.
99. Comparing multi-model mosaic and multi-model combination methods to simulate streamflow across the contiguous USA
Core Problem: Selecting one hydrological model per catchment may waste complementary process strengths, while fixed model averaging can fail as flow regimes change.
Key Innovation: Seventy-eight FUSE structures across 544 CAMELS catchments compare mosaic selection with static and dynamic multi-model combinations under joint high- and low-flow metrics.
100. Simulation and evaluation of local daily temperature and precipitation series derived by stochastic downscaling of ERA5 reanalysis
Core Problem: ERA5 is physically consistent but often too coarse to reproduce local precipitation and temperature sequences used in impact modelling.
Key Innovation: A GAM-ARMA stochastic downscaling framework reconstructs daily local series and improves multiple distributional and temporal diagnostics at more than 4,000 European stations over seven decades.
101. Satellite observations reveal widespread alteration of river thermal regimes by US dams
Core Problem: Evidence for downstream thermal alteration by dams has been dominated by individual rivers, seasons, or facilities, obscuring its national extent.
Key Innovation: Satellite temperature records compare reaches upstream and downstream of large US dams across seasons, revealing widespread and persistent shifts in river thermal regimes.
102. Metamorphic sulfur release as a driver of sustained cooling and mass extinction
Core Problem: Climate effects of large igneous provinces usually omit sulfur released when intrusions heat sedimentary host rocks because the emissions are not assumed to reach the stratosphere.
Key Innovation: Coupled metamorphic and climate modelling shows that sustained contact-metamorphic sulfur can generate centennial cooling spikes superimposed on longer carbon-dioxide warming.
103. Recent equatorward shift of the summer North Atlantic jet dominated by internal climate variability
Core Problem: The recent equatorward shift of the summer North Atlantic jet conflicts with the poleward zonal-mean trend, complicating attribution of associated extreme weather pathways.
Key Innovation: Reanalysis and model experiments identify internal variability in the North Atlantic warming hole as the dominant recent driver, with an externally forced signal projected to emerge around mid-century.
104. Global expansion of the sensitive aerosol-limited marine cloud regime under emission reductions
Core Problem: Aerosol-cloud forcing is most sensitive in aerosol-limited regimes, yet the geographic extent of that nonlinear state is changing as emissions decline.
Key Innovation: Satellite-model synthesis finds a 4.8% per decade global expansion over 20 years and projects two- to threefold increases in several ocean regions under strong emission reductions.
105. Numerical simulation on the thermal stability of highway embankments in the Qinghai-Tibet Plateau under different climate warming rates
Core Problem: The service life of permafrost highway embankments depends on warming rate, slope aspect, and whether passive cooling can prevent progressive thaw.
Key Innovation: Fifty-year simulations compare fill, insulation-board, and insulation-plus-thermosyphon designs under three warming rates, identifying where frozen ground persists and where higher rates overwhelm the cooling system.
106. UAV 3D Scene Understanding: A Survey from an Agent-Capability Evolution Perspective
Core Problem: UAV perception research is fragmented across mapping, online understanding, and predictive autonomy, making system capabilities difficult to compare.
Key Innovation: The review organizes UAV 3D scene understanding into offline interpretation, online understanding, and predictive reasoning, then identifies scene memory, collaborative fusion, sim-to-real transfer, and onboard deployment as the main unresolved transitions.
107. A Sub-Scene-Based GNSS-Constrained Structure from Motion for Robust Long-Corridor UAV Image Reconstruction
Core Problem: Weak long-corridor imaging geometry produces systematic bowl deformation in conventional UAV structure-from-motion reconstructions.
Key Innovation: Sub-scene stability tests, GNSS-constrained structureless bundle adjustment, and weighted global refinement reduce deformation to a mean 3D checkpoint RMSE of 0.051 m with one control point while cutting runtime by about 52% relative to COLMAP.
108. Intensifying Drought Under a Warming–Wetting Climate: Multi-Scale Impacts on Vegetation Phenology and Productivity in Xinjiang, China
Core Problem: Regional greening can coexist with intensifying soil-water stress, making climate trends alone insufficient for judging vegetation resilience in arid landscapes.
Key Innovation: Multi-scale remote sensing separates phenological and productivity responses to drought across Xinjiang under simultaneous warming and increasing precipitation.
109. Automated Victim Detection from UAV Thermal Infrared Imagery for Nighttime Search and Rescue Using Multi-Pose Ground Camera Data
Core Problem: Nighttime UAV search-and-rescue models face scarce aerial thermal labels, large viewpoint shifts, and strong target-scale changes with altitude.
Key Innovation: Ground-camera thermal images are transformed into aerial-like training views for Grounding DINO; external tests expose a specialization trade-off but show that altitude-aware rescaling recovers much of the scale-induced loss without retraining.
110. Filling Satellite Microwave Observation Gaps via Generative Synthesis
Core Problem: Polar-orbiting microwave radiometers leave multi-hour gaps during rapidly evolving weather, precisely when continuous all-weather observations are most valuable.
Key Innovation: A generative synthesis framework reconstructs the missing microwave sequence from surrounding observations, targeting temporal continuity without discarding the uncertainty inherent in unobserved states.
111. Multi-Scale Assessment of Nighttime Heat Health Risk and Dominant Factors Using MODIS and SDGSAT-1 Observations
Core Problem: Nighttime heat exposure varies below the scale resolved by conventional thermal-risk products, limiting identification of vulnerable urban neighbourhoods.
Key Innovation: MODIS and SDGSAT-1 observations are combined across scales to map nighttime heat-health risk and separate its dominant environmental and built-form controls.
112. Numerical analysis of the integrated hydrology of a hillslope regulated river basin
Core Problem: Mountain catchments with sparse observations require a single representation of overland flow, infiltration, recharge, groundwater, and abstraction.
Key Innovation: A transient finite-element model integrates field, satellite, hydromorphological, and GIS evidence to estimate coupled surface and groundwater discharge in Ecuador's Casacay catchment.
113. Interpretable machine learning: a comparison of TabNet and XGBoost for axial pile capacity prediction using layered soil representations
Core Problem: Axial pile-capacity models must transfer across sites with layered soils rather than fit repeated tests from the same location.
Key Innovation: Site-separated evaluation shows that TabNet generalizes better than an overfit XGBoost baseline and provides feature masks whose layered-soil contributions can be checked against geotechnical reasoning.
114. Oblique Pullout Resistance of Horizontal Ground Anchors in Frictional Slopes
Core Problem: Ground-anchor pullout resistance changes sharply near a slope crest, where inclination, embedment, and soil friction alter the available failure zone.
Key Innovation: A Mohr-Coulomb limit-equilibrium solution optimizes the failure surface and produces dimensionless resistance relations validated against existing anchor data.
115. UAV-derived bathymetry of a gravel-bed river mesohabitat mapping and hydromorphological assessment
Core Problem: River bathymetry needed for hydraulic and habitat models is expensive to survey continuously and is sensitive to both spectral model choice and mapping resolution.
Key Innovation: A Pareto-selected UAV workflow combines optical depth retrieval, structure-from-motion terrain, and calibrated 2D hydraulics to reconstruct gravel-bed morphology at two dates.
116. LLM-ResiGame: Multi-agent large language models for creating scenario-based resilience games in critical infrastructure decision-making practices
Core Problem: Static resilience exercises cannot adapt scenarios as infrastructure failures, cyberattacks, and natural hazards evolve during training.
Key Innovation: LLM-ResiGame assigns scenario generation and event management to multiple language-model agents, then evaluates scenario relevance, customization, and user engagement in an interactive study.
117. A 12-target global framework for measuring drought resilience: Insights from a multi-country review
Core Problem: Drought-resilience assessments lack a common set of measurable targets spanning basic needs, ecosystems, institutions, and crisis response.
Key Innovation: Review of 16 national policies yields 12 targets, 45 sub-targets, and 129 indicators organized into four operational resilience domains.
118. Shaking urban crime: Earthquakes, stress and crime in neighbourhoods
Core Problem: Post-earthquake changes in routine activity and social control may redistribute crime, but neighbourhood-scale spatial effects remain poorly documented.
Key Innovation: Local spatial statistics across 1,807 neighbourhoods identify new post-earthquake hotspots and offence-specific changes in the 196 areas affected by building collapse.
119. Census to sense: Decoding spatio-temporal dynamics of social vulnerability to disasters in Nepal
Core Problem: Composite vulnerability scores can conceal whether changing disaster risk reflects rising sensitivity or gains in adaptive capacity.
Key Innovation: Municipal census data from 2011 and 2021 separate Nepal's sensitivity and adaptive capacity, revealing a 35% national vulnerability decline and three spatially distinct transition pathways.
120. Grid resilience under compound extreme weather events: The case of a windstorm followed by a heat wave
Core Problem: Infrastructure designed for one extreme can remain vulnerable when a second hazard arrives before the system has recovered.
Key Innovation: Power-flow simulations on a synthetic Texas grid quantify how windstorm damage followed by heat-driven demand changes resilience in current and heat-expanded networks.
121. A framework for the assessment of dynamic urban climate resilience considering effective water management system
Core Problem: Urban water resilience changes over time as flood exposure, infrastructure condition, and management capacity interact.
Key Innovation: DEMATEL-ISM causal structure is coupled to a dynamic Bayesian network to update 43 resilience parameters for the Barak Valley water system under changing conditions.
122. First multi-sensor characterization of fire-emitted potassium spectral signatures from space in low-light and daytime conditions
Core Problem: Satellite fire products do not routinely distinguish flaming phase or combustion efficiency from broadband radiance alone.
Key Innovation: TROPOMI and OCO-2 resolve the paired potassium emission lines from wildfire under low light and local noon, while principal-component amplitude tracks fire light and radiative power.
123. Remote sensing of planetary boundary layer from ground and space: Structure, thermodynamics, and boundary layer clouds
Core Problem: Planetary-boundary-layer observations remain fragmented by instrument, retrieved quantity, and sparse coverage over oceans, polar regions, and remote land.
Key Innovation: The review aligns lidar, radar, radio occultation, hyperspectral sounding, and cloud observations around boundary-layer structure, thermodynamics, turbulence, and clouds, exposing where measurements are complementary or non-equivalent.
124. Live fuel moisture mapping in unburned chaparral areas of Southern California with MODIS and VIIRS
Core Problem: Live fuel moisture controls chaparral fire behaviour, yet field sampling is sparse and cross-sensor continuity is uncertain.
Key Innovation: A 2003-2022 MODIS model is bias-corrected and transferred to VIIRS, testing whether spatially continuous fuel-moisture estimates remain stable across the sensor transition.
125. High-resolution L-band VOD retrieval using GNSS-T-calibrated multi-sensor fusion and self-supervised learning
Core Problem: Operational L-band vegetation optical depth products are too coarse to resolve canopy-scale water-status heterogeneity.
Key Innovation: GNSS transmissometry calibrates a self-supervised fusion of Sentinel-1, optical, GEDI, and climate features to retrieve forest VOD at 30 m; medium-term climate variables dominate while SAR retains rain-event sensitivity.
126. Characterizing effects of tree mortality from wildfire and bark beetles using satellite observations of solar-induced chlorophyll fluorescence
Core Problem: The sensitivity of solar-induced chlorophyll fluorescence to different forest-mortality agents and severities has not been established.
Key Innovation: Satellite SIF separates wildfire and bark-beetle mortality responses and detects beetle-driven decline before visible mortality in aircraft surveys, outperforming standard vegetation indices for early stress detection.
127. National-scale tidal flat DEM reconstruction using optical satellite imagery
Core Problem: Rapid tidal-flat change and difficult field access prevent frequent, national-scale elevation surveys.
Key Innovation: An optical-satellite workflow reconstructs a national tidal-flat DEM from waterline and image evidence, providing topography at a scale not practical with repeated field surveys.
128. UAS ecophysiology: Imaging spectroscopy can map canopy leaf water potential in a diverse forest ecosystem
Core Problem: Forest drought vulnerability is difficult to map because canopy water potential is physiologically meaningful but sparsely measured in the field.
Key Innovation: Species-aware short-wave-infrared UAS spectroscopy predicts canopy leaf water potential before and after rehydrating rainfall, reaching strong field agreement across a diverse forest.
129. Thermodynamically constrained surface energy balance using medium-resolution remote sensing for efficient evapotranspiration mapping
Core Problem: Medium-resolution evapotranspiration models often fail outside croplands because aerodynamic parameterization does not transfer across surface-energy regimes.
Key Innovation: RADET uses a diffusivity-independent equilibrium formulation with conditional advective enhancement and evaluates Landsat-based estimates across 145 flux stations and multiple land-cover types.
130. Towards physically consistent soil roughness for CIMR L-band soil moisture and VOD retrievals
Core Problem: The fixed viewing geometry of the forthcoming CIMR mission requires a soil-roughness calibration that cannot be transferred directly from multi-angle microwave sensors.
Key Innovation: A physically consistent H-Q-N parameterization is calibrated from six years of mono-angular SMOS observations to support continuity in L-band soil-moisture and vegetation-optical-depth retrieval.
131. Improved mapping of Arctic fractional land cover and land cover change from multi-resolution optical remote sensing
Core Problem: Arctic shrub expansion is difficult to quantify because vegetation classes are spectrally similar, ground data are sparse, and 30 m pixels mix multiple covers.
Key Innovation: Commercial 2 m imagery trains annual 30 m fractional-cover maps from Landsat and Sentinel-2 for 2016-2023 across 335,000 km2, with pixelwise uncertainty and independent change validation.
132. A causal inference-guided dual-contrast decoupling framework for long-tailed SAR target recognition
Core Problem: Real SAR training sets are long-tailed because both class frequency and scene context bias the observed target distribution.
Key Innovation: A structural causal model separates intrinsic class imbalance from scene-induced spurious imbalance, then applies dual contrastive objectives to learn less biased target representations.
133. A source-free unsupervised domain adaptation framework for large-scale, in-season soybean mapping
Core Problem: In-season crop maps transfer poorly across regions because phenology, management, and cloud sampling shift without providing new labels.
Key Innovation: CROPS adapts a pretrained soybean segmenter using only unlabeled target imagery, combining source-free contrastive representation learning with optimized class prototypes.
134. MASt3R-Fusion: Integrating feed-forward visual model with IMU, GNSS for high-functionality SLAM
Core Problem: Feed-forward visual geometry is strong in difficult scenes but lacks the scale, temporal consistency, and uncertainty control supplied by inertial and GNSS measurements.
Key Innovation: MASt3R-Fusion tightly couples learned point maps with IMU and GNSS constraints in a probabilistic SLAM system, retaining metric consistency when image-only estimation degrades.
135. Open-source optical-radar data fusion for mapping invasive tree species: A scalable framework for regional LULC applications
Core Problem: Invasive trees are poorly represented in regional land-cover products despite their management importance and variable optical appearance.
Key Innovation: An open-source workflow fuses optical and radar time series to map invasive Acacia at regional scale and exposes the full processing chain for reuse in other LULC applications.
136. Integrating cloud motion information: A deep learning approach for real-time wind field retrieval using geostationary satellite imagery
Core Problem: Satellite wind retrieval often treats images independently and discards cloud motion that directly encodes atmospheric advection.
Key Innovation: CMW_SimVP extracts motion from three Himawari-8 images, improving 10 m wind accuracy by more than 15% and producing a 10-minute, 0.25-degree field in under one second on one GPU.
137. STAR-IOD: Scale-decoupled topology alignment with pseudo-label refinement for Remote Sensing Incremental Object Detection
Core Problem: Incremental remote-sensing detectors forget old classes when object scale changes and old objects are unlabeled in new training batches.
Key Innovation: STAR-IOD aligns class topology in scale-decoupled subspaces and generates cluster-calibrated pseudo-labels; it also releases DIOR-IOD and DOTA-IOD benchmarks and improves mAP by 1.7-2.1 points.
138. Towards generalized spaceborne SAR ship detection via Fourier-based perturbation augmentation and quality-aware invariance learning
Core Problem: Spaceborne SAR ship detectors trained on one platform degrade on unseen sensors because image spectra and quality shift.
Key Innovation: FAQA perturbs Fourier-domain appearance during training and enforces quality-aware invariance, targeting cross-platform detection rather than sensor-specific accuracy.
139. A data- and knowledge-driven cropland parcel recognition method based on segment anything model (SAM)
Core Problem: Cropland parcel mapping needs boundary labels that are costly to draw and difficult to keep consistent across heterogeneous imagery.
Key Innovation: Segment Anything generates parcel candidates from multisource imagery and land-survey data, after which a multitask decoder jointly refines parcel extent and boundaries.
140. Drought effect on spectral response and burn assessments in tropical environments
Core Problem: Burn-severity thresholds derived in temperate forests may fail where tropical vegetation is already dry before fire and therefore has a low pre-fire near-infrared baseline.
Key Innovation: Sentinel-2 spectra from five tropical environments show that NBR can detect burns while standard dNBR thresholds systematically understate severity, including classifying clearly separable scars as unburned.
141. Heterogeneous vegetation resilience trajectories beneath greening during open-pit mine reclamation
Core Problem: Apparent vegetation greening after open-pit mining can mask unstable recovery trajectories and recurrent disturbance.
Key Innovation: VegDecouple separates background and short-term Landsat NDVI dynamics from 1990 to 2025, exposing heterogeneous resilience beneath reclamation greening on the Loess Plateau.
142. Physically consistent multitemporal dual-domain learning for dual-polarization SAR despeckling
Core Problem: Real dual-polarization SAR despeckling lacks clean targets and must distinguish speckle from genuine temporal and polarimetric change.
Key Innovation: CD2P-Net learns jointly in spatial and transform domains, using neighbouring acquisitions and scattering consistency to suppress speckle without erasing structural or polarization information.
143. Quantifying the influence of DEM elevation errors on regional gravity forward modeling through DEM correction and fusion
Core Problem: DEM elevation errors propagate directly into high-frequency gravity forward models, but correction and uncertainty transmission are rarely evaluated together.
Key Innovation: An integrated DEM correction-and-fusion workflow quantifies how terrain bias changes regional gravity estimates before and after elevation adjustment.
144. Improving urban catchment delineation through comparative DEM accuracy assessment and bias correction techniques
Core Problem: Surface-biased global DEMs can displace drainage divides and corrupt urban catchment boundaries, especially among buildings and engineered relief.
Key Innovation: Comparative error analysis and bias correction across global DEMs show how vertical adjustment improves watershed delineation and hydrological interpretation in an urban setting.
145. The Brazilian soil total available water (TAW) acquired by remote sensing and machine learning techniques
Core Problem: National soil available-water mapping is constrained by sparse profiles and uncertain conversion from soil texture to hydraulic storage.
Key Innovation: Pedotransfer estimates from 41,438 profiles and a random forest using spectral, soil, and terrain covariates produce a 30 m Brazilian available-water map with explicit validation.
146. A theoretical model of the influence of physical soil crusts on wind erosion under freeze-thaw action
Core Problem: Wind-erosion models omit the loss of soil-crust cohesion caused by repeated freezing and thawing.
Key Innovation: A particle moment-balance parameterization links moisture, crust hardness, freeze-thaw cycles, wind speed, and grain size to threshold velocity and erosion rate, with wind-tunnel validation.
147. Investigating impacts of soil drought propagation on vegetation restoration at multilayer scales on the Loess Plateau, China
Core Problem: Large-scale restoration on the Loess Plateau changes vegetation demand, but drought propagation through the upper two metres of soil remains poorly resolved.
Key Innovation: Multi-depth soil-moisture indices show rapid shallow droughts, persistent deep deficits, and strongest cumulative vegetation response to drought in the 10-100 cm layer.
148. A novel 3D geospatial framework for modeling lateral and vertical heterogeneity in complex soil deposits: A unified stratigraphic approach
Core Problem: Two-dimensional geotechnical maps oversmooth depth-dependent heterogeneity and cannot preserve cross-axis covariance in complex deposits.
Key Innovation: A vertically extended empirical-Bayesian-kriging framework constructs full 3D soil-property maps, reducing error by as much as 75% and retaining substantially more fine-scale variance than baseline methods.
149. Multi-scale contact characteristics of rock joints under normal loading: Integrating asperity and system scales
Core Problem: Rock-joint contact area, stiffness, and stress depend on roughness at both asperity and system scales, limiting extrapolation from laboratory specimens to slopes or underground openings.
Key Innovation: Boundary-element simulations with plasticity derive a unified scaling law for contact patches and separate how peak geometry, rock hardness, and scale govern normal loading.
150. Characterizing spatiotemporal drought dynamics in the Yangtze River Basin: from event lifecycle to shifting climate drivers
Core Problem: Yangtze drought assessments rarely track complete event lifecycles or separate changing precipitation supply from evaporative demand.
Key Innovation: Three-dimensional event clustering reconstructs drought initiation, expansion, migration, and termination, then attributes changing behaviour to shifting climate drivers.
151. Global cropland expansion amplifies the risk of compound dry and hot events under climate change
Core Problem: Global projections of compound dry-hot exposure often hold cropland extent fixed and therefore miss the risk added by agricultural expansion.
Key Innovation: CMIP6 projections and dynamic cropland maps partition climate- and land-expansion contributions across five severity classes, identifying the largest exposure increases in South America, Sub-Saharan Africa, and Asia.
152. HYDRO-XM: a computationally efficient, externally coupled, and flexible watershed model
Core Problem: Watershed models often splice surface runoff, vadose-zone flow, and groundwater using fixed boundary assumptions that fail during rapid wetting and saturation.
Key Innovation: HYDRO-XM couples KINEROS2, HYDRUS-1D, and MODFLOW with dynamic boundary switching, adaptive time steps, and parallel execution across the surface-subsurface continuum.
153. Determining and predicting single-storm erosion index in Sicily
Core Problem: Single-storm erosivity requires short-interval rainfall intensities that are unavailable at many gauges, and historical depth-based equations may not remain stationary.
Key Innovation: Analysis of 8,464 Sicilian storms quantifies the loss from 30-minute sampling, improves depth-to-erosivity estimation, and shows that 1951-1970 relations misrepresent 2002-2024 events.
154. Dynamic changes of vegetation in China under the stress of compound droughts
Core Problem: Vegetation exposure to compound high vapour-pressure deficit and low soil moisture lacks a unified metric for long-term regional comparison.
Key Innovation: A standardized compound-drought index and random-forest attribution show increasing compound stress across China and quantify where carbon-dioxide fertilization fails to offset it.
155. Tailoring soil moisture data to diverse accuracy needs: a multi-objective optimization and point-surface fusion framework
Core Problem: Soil-moisture fusion methods usually optimize one error metric and impose restrictive independence assumptions, despite application-specific accuracy needs.
Key Innovation: A Bayesian multi-objective point-surface framework combines four satellite products, three land models, and ground observations to tune complementary accuracy objectives.
156. Urbanization is associated with changes in the sensitivity of water quality to drought
Core Problem: Urbanization changes baseline river chemistry and may also alter how water quality responds to drought-related low flow.
Key Innovation: Regression across urbanization gradients separates baseline shifts from discharge sensitivity and resolves contrasting seasonal responses in conductance and water temperature.
157. A mechanistic water erosion-carbon model integrating the soil aggregate fractal dimension
Core Problem: Erosion-carbon models can mistake concentration enrichment after fine-particle removal for a gain in soil carbon stock.
Key Innovation: A fractal aggregate model and flume tests enforce mass balance, showing apparent enrichment at low flow but 25-30% organic-carbon loss after aggregate breakdown at higher intensity.
158. A unified framework for groundwater level imputation and forecasting in data-limited catchments
Core Problem: Groundwater forecasting in sparse networks must remain reliable when much of the training record has been reconstructed rather than observed.
Key Innovation: GWL-IMFO combines chained-equation imputation with neural architecture search and stress-tests 20-40% missing training data across 1,493 wells while retaining 2024 for independent evaluation.
159. A deep learning-based probabilistic framework for streamflow prediction and water quality assessment
Core Problem: River water-quality forecasts inherit uncertain inflows, but deterministic hydrological inputs conceal the probability of regulatory exceedance.
Key Innovation: LSTM streamflow ensembles are spatially disaggregated into a process-based water-quality model, reproducing BOD, phosphorus, and ammonium status while assigning probabilities to threshold exceedance.
160. Enhancing water management in data-scarce watersheds using satellite and reanalysis precipitation: a combined SWAT+ and SMOGN machine learning approach
Core Problem: Data-scarce watersheds lack a clear basis for choosing among satellite and reanalysis precipitation products when both high and low flows matter.
Key Innovation: SWAT+ and SMOGN-assisted machine-learning models compare MSWEP, IMERG, and ERA5-Land using conventional scores and flow-duration curves, exposing product performance across the full flow regime.
161. Intelligent mining of generation operation rules for cascade hydropower reservoirs driven by large language models and knowledge graphs
Core Problem: Cascade-reservoir operating rules are dispersed across unstructured literature and difficult to reconcile with flood-control, ecological, and power constraints.
Key Innovation: A language model extracts rules into a domain knowledge graph before constrained optimization, producing interpretable operating policies for four lower Jinsha River reservoirs.
162. Random Fourier mapping physics-informed neural networks based digital volume correlation for full-field 3D deformation measurement of geomaterials
Core Problem: Digital volume correlation amplifies noise during strain differentiation and struggles with high-frequency internal deformation in rocks and soils.
Key Innovation: Random Fourier features expand a physics-informed DVC model so full-field 3D displacement and strain remain stable under complex geomaterial deformation.
163. Knowledge-informed data-efficient characterisation of soil spatial variability
Core Problem: Sparse, expensive site investigations limit reliable mapping of quasi-site-specific soil variability.
Key Innovation: Knowledge-informed multifidelity learning uses active selection of high-fidelity tests and transferable low-fidelity profiles, matching baseline accuracy with half the expensive target-site data.
164. Prior-guided online estimation of peak proximity from partial shear-curve geometry in multi-stage direct shear tests
Core Problem: Automated multi-stage direct-shear tests need to stop near the peak before the complete stress-displacement curve is available.
Key Innovation: A two-layer supervisory estimator uses partial curve geometry and prior stage information to infer peak proximity online, evaluated on 297 laboratory tests across soil types and states.
165. Fast 3D diffusion for scalable discrete granular media synthesis
Core Problem: Preparing stable initial particle assemblies can consume days before a large discrete-element simulation begins.
Key Innovation: Unconditional 3D diffusion generates local granular volumes and an inpainting model stitches them into arbitrarily large DEM-compatible assemblies, reducing example initialization from days to hours beyond 200,000 particles.
166. Ultra-low frequency solid-liquid topological seismic metamaterial multi-band rainbow capture and robust transmission for water environment
Core Problem: Conventional seismic metamaterials occupy land and often attenuate only narrow or high-frequency bands.
Key Innovation: A solid-liquid topological design places multiple robust interface modes below 20 Hz in a water environment and uses graded cells for multi-band wave localization.
167. Unconfined compressive strength prediction of cement-stabilized Aeolian sand under compaction-induced time delay effects using physics-informed neural networks (PINNs)
Core Problem: Cement-stabilized aeolian sand has a non-monotonic strength response to compaction delay, but small experimental datasets make unconstrained prediction unreliable.
Key Innovation: Physics-informed descriptors and a delay-based neural ODE encode hydration gain and workability loss, reaching cross-validated R-squared of 0.96 with conformal uncertainty coverage.
168. ‘A flash in the pan’ or a more enduring contribution? A retrospective on Micha Klein's (1976) ‘Hydrograph peakedness and basin area’, Earth Surface Processes, vol. 1: 27-30
Core Problem: The hydrograph peakedness index introduced in 1976 is rarely used today, leaving its physical interpretation and value for modern fluvial comparison uncertain.
Key Innovation: The retrospective reconstructs the index, reviews what subsequent basin studies learned from it, and tests whether a scale-normalized measure of mean-to-peak flow still adds information to flood hydrology.
169. A Novel Sequential Monte Carlo Algorithm for Parameter Estimation in Eco-Hydrological Models
Core Problem: Bayesian calibration of high-dimensional ecohydrological models is limited by multimodal posteriors and the cost of repeated model evaluation.
Key Innovation: PATPEMS combines adaptive sequential Monte Carlo, particle rejuvenation, reflective boundaries, and parallel evaluation to preserve posterior exploration while reducing wall-clock time.
170. ADA-IRSTD: Active Domain Adaptation for Cross-Domain Infrared Small Target Detection
Core Problem: Infrared small-target detectors trained on synthetic or narrow datasets degrade in real scenes, yet full target-domain annotation defeats the purpose of adaptation.
Key Innovation: ADA-IRSTD masks background while preserving target context, predicts false-alarm risk, and selects globally uncertain target samples to concentrate annotation on the most informative images.
171. MVTopo: Multiview Topology-Aware Prototype Reconstruction for Generalized Category Discovery in Remote Sensing Images
Core Problem: Generalized category discovery in remote sensing must retain known classes while separating unseen classes despite spectral similarity, intraclass variation, and noisy boundaries.
Key Innovation: MVTopo constructs semantic, spectral, and spatial prototypes and weights neighbours by manifold-topology confidence, reducing boundary-driven false assignment of novel categories.
172. Bistatic SAR Target Recognition Using Transferred Monostatic Knowledge and Cross-Domain Feature Fusion
Core Problem: Bistatic SAR recognition is constrained by scarce labelled observations even where abundant monostatic imagery exists for the same target classes.
Key Innovation: A paired monostatic-bistatic electromagnetic simulation dataset supports adversarial transfer and attention-based cross-domain fusion, raising three-class ship recognition to a 68.08% F1 score.
173. The Cooling Efficiency Factor Index (CEFI): A New Satellite-Based Dataset for Research and Operational Monitoring of Land Surface Processes
Core Problem: Land-surface cooling efficiency varies through turbulent and ecohydrological fluxes but is difficult to observe consistently in near real time.
Key Innovation: CEFI derives daily apparent heat-capacity change from geostationary satellites at 5 km from 2005 onward, providing an operational proxy for evapotranspiration, drought stress, fire risk, and wind anomalies.
174. ClimateBenchPress (v1.0): a benchmark for lossy compression of climate data
Core Problem: Lossy compression is becoming unavoidable for petabyte-scale climate output, yet compressors cannot be compared without tests of whether scientific structure survives.
Key Innovation: ClimateBenchPress defines common model and satellite datasets, variable-specific error bounds, and scientific evaluation tests for reproducible comparison of climate-data compressors.
175. PanDiM: A Diffusion Mamba Network for High-Fidelity Pansharpening
Core Problem: Diffusion pansharpening uses fixed denoising dynamics even though degradation and reconstruction needs change across space, frequency, and diffusion time.
Key Innovation: PanDiM predicts high-frequency residuals with degradation-posterior guidance, time-aware Mamba regulation, and frequency-decoupled loss, outperforming comparison methods on WorldView-3, GaoFen-2, and QuickBird.
176. Intraspecific Drought Stress Responses Detected Using Remotely Sensed Vegetation Indices in Sitka Spruce (Picea sitchensis)
Core Problem: Species-level drought response can be obscured when remotely sensed vegetation indices are interpreted as a uniform forest signal.
Key Innovation: The study evaluates multiple vegetation indices for detecting within-species Sitka spruce drought stress and clarifies which spectral responses remain diagnostic under changing conditions.
177. Earth Observation Data and Indigenous Perspectives: Two-Eyed Seeing Approach to Understanding Long-Term Wildfire and Landscape Changes
Core Problem: Wildfire and rapid warming on Indigenous northern lands cannot be understood fully from satellite trends without the environmental history held by local communities.
Key Innovation: Earth observations are interpreted with Tlicho Elders' knowledge through Two-Eyed Seeing, linking forest conversion, expanding shrub and grass cover, warming, snow decline, and drying soils to a reinforcing wildfire-risk cycle.
178. Forest disturbance detection and spatio-temporal dynamic analysis using the GRU-LPETransformer model
Core Problem: Forest-disturbance time series must capture abrupt local change and long temporal dependence while retaining the year of disturbance.
Key Innovation: GRU-LPETransformer combines recurrent local dynamics with position-aware global attention, reaching 82.96% F1 for disturbance and 74.14% disturbance-year accuracy within one year in Fujian.
179. Towards a mitigation behavioral framework for indoor wildfire smoke: The response to indoor pollution exposure (RIPE) model
Core Problem: Wildfire-smoke advice is difficult to translate into household action because perceived exposure, indoor conditions, and feasible mitigation interact.
Key Innovation: The RIPE framework organizes behavioural response to indoor wildfire-smoke exposure and identifies the conditions under which protective actions are likely to be adopted.
180. Stakeholder perceptions of effective management of disaster-generated debris after Hurricanes Helene and Milton in Florida, USA
Core Problem: Post-hurricane debris plans often omit the operational constraints faced by public agencies, haulers, and monitors during consecutive events.
Key Innovation: Thirty-two stakeholder interviews after Hurricanes Helene and Milton identify equipment, labour, trucking, and pre-permitted staging capacity as recurrent recovery bottlenecks.
181. Hcropland30: A hybrid 30-m global cropland map generated by leveraging global land cover products and Landsat data via deep learning
Core Problem: Existing 10-30 m global cropland products disagree spatially, limiting reliable agricultural exposure and land-change analysis.
Key Innovation: Hcropland30 combines globally stratified samples, ensemble-derived pseudo-labels, Landsat phenology, and a U-Net to map 2020 cropland at 30 m with 89.8% overall accuracy and 0.85 cropland F1.
182. SNSTFM: A novel data fusion approach based on spatial non-stationarity to generate fine-resolution land surface temperature
Core Problem: Fine-resolution land-surface-temperature fusion is biased when temporal relationships vary spatially rather than remaining stationary across the scene.
Key Innovation: SNSTFM estimates location-dependent cross-scale relationships to fuse coarse frequent observations with fine sparse imagery while retaining local thermal structure.
183. STaPL: Scale Transfer with Pseudo-Labelling for satellite-based mapping of agricultural practices
Core Problem: Agricultural conservation practices are reported only at coarse administrative scales, while field labels needed by conventional satellite models are unavailable for most places and years.
Key Innovation: STaPL converts county statistics into iteratively refined field pseudo-labels, mapping tillage from Landsat with 75% accuracy and stronger temporal transfer than field-trained references.
184. UAV-satellite synergies for crop monitoring: current landscape, challenges, and future directions
Core Problem: UAV and satellite crop observations are difficult to combine because spatial, temporal, and signal mismatches undermine model transfer.
Key Innovation: The review organizes sensors, fusion strategies, and applications, then identifies adaptive multimodal alignment and autonomous multi-platform acquisition as the main development needs.
185. Detecting intact forests in a highly fragmented landscape: Complementary insights from Landsat and GEDI
Core Problem: Fragmented landscapes lack an operational definition of intact forest that combines long-term canopy stability with present-day structure.
Key Innovation: A Landsat-LandTrendr workflow identifies fragments with near-zero 1984-2023 canopy change, while GEDI and field plots test structural separation of intact and altered forest at footprint and fragment scales.
186. TAIS-Net: Time adaptive implicit sampling diffusion model for arbitrary-scale UAV video super-resolution
Core Problem: UAV video super-resolution must handle large camera motion, weak texture, temporal drift, and user-selected scale in one reconstruction.
Key Innovation: TAIS-Net aligns prior frames with optical flow, uses implicit diffusion for arbitrary-scale detail, and corrects latent history during sampling to suppress flicker and accumulated error.
187. C2D2: Matching UAV images with cycle-consistency and channel-separation for joint detector and descriptor
Core Problem: Joint UAV keypoint detection and description suffer from competing feature requirements and patch losses that ignore global geometric consistency.
Key Innovation: C2D2 separates task channels and adds cycle consistency, improving detector repeatability and day-night localization with a 0.49-million-parameter model.
188. Spatial and temporal patterns of early leaf discoloration in Switzerland’s European beech forests
Core Problem: Drought-induced early leaf discoloration is confounded by seasonal phenology and multi-year legacy effects in satellite time series.
Key Innovation: Field observations and normalized Sentinel-2 indices isolate early discoloration and map beech-forest stress across Switzerland at 10 m from 2017 to 2023.
189. Global assessment of climate-driven lag and cumulative effects on vegetation using Earth Observation data and multivariable machine learning models
Core Problem: Global vegetation models commonly treat climate response as instantaneous, overlooking lagged and cumulative effects that vary by climate zone.
Key Innovation: Earth-observation time series and multivariable machine learning map those temporal effects globally and show that including lag terms materially changes regional model skill and driver attribution.
190. MS-AFNet: A multi-scale adaptive fusion network with high-resolution feature enhancement for tiny object detection in UAV imagery
Core Problem: Tiny targets in UAV imagery lose spatial detail through downsampling and are easily confused with clutter.
Key Innovation: MS-AFNet retains a high-resolution branch, adaptively fuses scales, and improves VisDrone test-dev AP50 by 7.7 percentage points over YOLO11n while remaining compact.
191. A global 500 m dataset of tree crown morphology to advance vegetation modeling and management
Core Problem: Tree-crown depth and width shape ecosystem exchange but remain globally unmapped because field and LiDAR observations are sparse.
Key Innovation: More than 225,000 crown observations train plant-functional-type-specific models that produce 500 m global crown depth, width, aspect ratio, and uncertainty layers, substantially outperforming allometric equations.
192. Synergizing greening and resilience: a remote sensing-driven spatial optimization framework based on landscape configuration
Core Problem: Vegetation expansion does not guarantee ecological resilience when the spatial arrangement of restored cover is poorly configured.
Key Innovation: Interpretable remote-sensing analysis informs a landscape optimizer that triples resilience relative to random greening and identifies diminishing returns beyond a 7.6% forest-expansion threshold in the study region.
193. Quantifying fragile carbon balance by multiple remote sensing based vegetation optical depth datasets
Core Problem: Net tropical carbon balance can conceal large opposing fluxes from disturbance and recovery, leading stable totals to be mistaken for stable forests.
Key Innovation: L-band vegetation optical depth, high-resolution disturbance maps, and nine vegetation models separate carbon gains and losses across tropical America from 2011 to 2020.
194. GlobalWR-2025: The first global 10-meter winter rapeseed dataset developed by knowledge-guided temporal features
Core Problem: A globally consistent 10 m winter-rapeseed map has been blocked by sparse labels, threshold sensitivity, and regional phenological differences.
Key Innovation: WROS encodes optical-SAR phenology into transferable temporal features and produces GlobalWR-2025, mapping about 34.74 million ha with overall accuracy above 92% and independent regional accuracy above 93%.
195. Benefits of high-resolution satellite evapotranspiration for agro-silvo-pastoral systems in the Sahel
Core Problem: Coarse evapotranspiration products smooth the surface heterogeneity that controls drought monitoring in Sahelian agro-silvo-pastoral mosaics.
Key Innovation: Comparison of 1 km MODIS and 30 m Landsat energy-balance estimates shows where finer thermal observations reduce uncertainty and recover species- and land-use-scale water patterns.
196. Assessing UAV imagery and high-resolution LiDAR for tree height estimation: The role of flight speed and image overlap
Core Problem: UAV tree-height estimates depend on flight speed and image overlap, but those acquisition choices are rarely evaluated jointly against LiDAR and extensive field measurements.
Key Innovation: A factorial survey of 12 flight configurations and 920 trees quantifies the accuracy, processing, and mission-efficiency trade-offs of structure-from-motion and LiDAR canopy models.
197. Comparison of GPM IMERG V07B versus IMERG V06B Precipitation in Northern Spain and their application for wind turbine blade lifetime estimation
Core Problem: Satellite precipitation updates must be tested across rainfall phase, intensity, and complex terrain rather than by aggregate totals alone.
Key Innovation: Twenty-eight gauges in northern Spain show where IMERG V07B improves frozen and mixed precipitation over V06B and where sub-daily bias remains.
198. Three-dimensional chlorophyll content retrieval in forest canopies via physics guided transfer learning and UAV multi-source remote sensing
Core Problem: Sparse field labels limit three-dimensional retrieval of canopy chlorophyll from combined UAV spectral and structural observations.
Key Innovation: Physics-guided transfer learning pre-trains on radiative-transfer simulations and fine-tunes on field data, outperforming lookup-table, hybrid, and purely data-driven baselines for vertical chlorophyll profiles.
199. Responses of shallow soil respiration and moisture to climate warming in permafrost regions of the Qinghai-Tibet plateau: a meta-analysis
Core Problem: Permafrost warming alters coupled soil respiration and moisture, but evidence varies across depth, vegetation, and observation period.
Key Innovation: A meta-analysis synthesizes shallow-soil thermal, moisture, and carbon responses across Qinghai-Tibetan permafrost studies to separate consistent effects from site dependence.
200. Soil moisture dynamics and enhancement of satellite-derived soil moisture responses to rainfall events across land use and land cover types
Core Problem: Satellite soil-moisture response to rainfall depends on land cover and antecedent storage, complicating interpretation of the same retrieval across landscapes.
Key Innovation: The study compares event-scale moisture dynamics among land-use classes and quantifies where satellite-derived signals amplify or damp rainfall response.
201. Shifts from climate-driven to human-influenced fire regimes in central China over the last two millennia
Core Problem: Modern fire observations are too short to distinguish climatic control from long-term human reshaping of regional fire regimes.
Key Innovation: A two-millennia record from central China identifies the transition from climate-dominated variability to stronger anthropogenic influence on burning.
202. From environmental filtering to stochastic assembly: An operational framework for post-fire soil microbial succession
Core Problem: Post-fire soil microbial recovery reflects both deterministic environmental filtering and stochastic recolonization, but their relative importance changes through succession.
Key Innovation: An operational assembly framework traces that transition and identifies the environmental and network attributes governing recovery after fire.
203. Vertically increasing rock fragment content exacerbates drought stress in Dolomite-derived soils of rocky desertification areas
Core Problem: Rock fragments alter infiltration, evaporation, and heat transfer in shallow dolomitic soils, but their effects depend nonlinearly on size, abundance, and vertical arrangement.
Key Innovation: Controlled reconstructed-soil experiments separate those interactions and show how layered, fragment-rich profiles intensify drought stress in rocky-desertification terrain.
204. Incorporating variable source areas into rainfall-runoff pollutant export modeling for rural residential catchment
Core Problem: Uniform contributing-area assumptions misrepresent the onset and peak of pollutant export from heterogeneous rural catchments during rainfall.
Key Innovation: A variable-source-area wash-off model adds parameters for maximum contributing extent and activation rate and improves dissolved nitrogen and phosphorus simulation in a Yangtze catchment.
205. Diverging seasonal trends in soil moisture in Northern Europe
Core Problem: Regional soil-moisture trends can reverse by season even when annual means appear stable, obscuring drought and recharge change.
Key Innovation: A Northern European analysis separates seasonal trajectories and their climatic controls rather than averaging opposing wetting and drying signals.
206. Three-dimensional reservoir characterization from hard and soft data based on latent diffusion model
Core Problem: Three-dimensional subsurface models must honor exact borehole values while using uncertain seismic interpretation to constrain structures between wells.
Key Innovation: A latent-diffusion model fuses hard and soft constraints through multi-level conditioning, reducing generative complexity while preserving geological continuity.
207. A decentralized multi-agent framework for adaptive water allocation in river-reservoir systems
Core Problem: Centralized water-allocation optimization poorly represents autonomous users and is difficult to coordinate across large river-reservoir systems.
Key Innovation: A distributed-constraint model compares four multi-agent algorithms and links seasonal forecasts to operating constraints, with the strongest methods approaching centralized cost while preserving supply and environmental flows.
208. Satellite water quality monitoring of China’s coastal inland waters based on Google Earth Engine
Core Problem: Large-area inland-water retrieval must combine optical evidence with environmental context without making continental processing prohibitive.
Key Innovation: A Google Earth Engine workflow fuses Sentinel-2 reflectance, station observations, and geographic covariates to retrieve four water-quality parameters across China's coastal inland waters.
209. 3D hydro-mechanical durability of lime-stabilised expansive marly clay pavement subgrades under wet-dry cycling
Core Problem: Laboratory wet-dry durability tests do not capture the three-dimensional hydro-mechanical response of stabilized pavement subgrades under field drought and traffic.
Key Innovation: A coupled finite-element model scales a standardized cycling protocol to full pavement geometry and compares untreated, granular-capped, and lime-stabilized structures through repeated drying, wetting, and loading.