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

TerraMosaic Daily Digest: August 30, 2026

August 30, 2026 TerraMosaic Daily Digest

Daily Summary

Direct geohazard papers concentrate on how instability emerges from interacting hydrologic, geomorphic, and material controls rather than from nominal triggers alone. Landslide studies separate common from type-specific controls in the Rif Mountains, show that weighted fusion of medium-resolution DEMs can approach LiDAR-referenced susceptibility performance in North Carolina, and recast landslide-dam hazard around inflow, storage, erosion, and water-balance processes. Rockfall analysis further indicates that small precursory failures are mechanically clustered rather than random, yet do not consistently accelerate pre-failure deformation. Engineering-scale slope papers add explicit temporal structure: anchored slopes show staged reliability decline with a transition near 30 years, while liquefiable sheet-pile systems depend on shifts between cyclic mobility and residual deformation accumulation rather than triggering metrics alone.

A second direct cohort resolves deformation and event response through distributed, multiscale observation. In Taiwan, typhoon-season bare-ground erosion rose sharply after Typhoon Krathon, with slope and vegetation cover dominating control; in eastern Afghanistan, field observations and PSInSAR indicate that the 2025 Asadabad-Kunar earthquake deformed a corridor wider than mapped fault traces and coincided with landslides and rock avalanches; and in the Natron Rift, shear-wave splitting identifies shallow crack-controlled anisotropy overprinted by mid-crustal magma storage and temporally variable stress rotation. Urban and infrastructure studies apply similar logic: Zhengzhou deformation separates into groundwater-linked uplift and oscillation plus construction-linked local subsidence, while long-term Shanghai railway monitoring isolates regional subsidence from acute construction settlement and cumulative consolidation.

A parallel methods stream expands operational observation and transferable inference, but often outside explicit geohazard validation. Hazard-facing examples include SAR detection of small avalanches, cross-disaster satellite damage classification, parcel-linked street-view recovery assessment after Hurricane Helene, flood platforms that connect segmentation to exposure analytics, precipitation assimilation that improves severe-rain forecasting, and heat or riverbank studies that extend risk analysis to transit microclimates and mental-health burden. More general remote-sensing and mechanics papers advance multisource visual prompting, spectral reconstruction and classification, lightweight detection, soil-moisture downscaling, cropland change detection, ocean-current inversion, scene segmentation, poroelastic wave simulation, nonstationary soil uncertainty, hydrophobic seepage monitoring, wildfire-altered soil geotechnics, and granular-fluid constitutive transitions; these enlarge the analytical toolkit, but they are not yet geohazard validations in themselves.

Key Trends

Five trajectories organize the August 30 literature: process-specific slope controls, distributed deformation diagnosis, consequence-oriented disaster intelligence, land-change-conditioned hydroclimatic risk, and structured transferable Earth-observation and mechanics methods.

  • Slope and landslide hazards are being resolved through hydrologic, type-specific, and lifecycle controls: The landslide cohort moves beyond undifferentiated susceptibility surfaces. Hydrologic review of landslide dams, multivariate separation of slides, flows, falls, and complex failures, DEM-fusion susceptibility mapping, rockfall precursor testing, anchored-slope degradation analysis, and liquefaction regime modeling all treat hazard behavior as conditional on process class, evolving state, and maintenance horizon.
  • Deformation analysis is shifting from mapped traces to distributed multiscale fields: Across the Natron Rift, the Asadabad-Kunar earthquake, Zhengzhou, Shanghai rail corridors, and typhoon-affected Taiwan, the key methodological move is to resolve deformation as laterally distributed, depth-dependent, or scale-separated structure. Seismic anisotropy, PSInSAR, SBAS-InSAR, wavelet decomposition, ICA, and residual-based anomaly detection are being used to distinguish regional background signals from local hazard concentrations.
  • Operational disaster intelligence is extending from detection to attribution, exposure, and recovery: Several papers no longer stop at delineating damaged or flooded pixels. Cross-disaster building damage assessment, AIoT-enabled flood impact mapping, Shanghai railway anomaly diagnosis, and Recov-Vision's parcel-scale occupancy inference all link remote observations to consequence, exposure, or maintenance decisions, while keeping uncertainty and transferability as explicit constraints.
  • Hydroclimatic risk studies are coupling land-change geometry with forcing and exposure: Port-au-Prince urban expansion, typhoon-driven erosion in Taiwan, FY-3G precipitation assimilation, SMAP downscaling, irrigation-sensitive humid heat metrics, transit-stop heat mapping, and vegetation productivity analysis each connect environmental forcing to spatial organization of runoff, moisture, or heat exposure. The common trajectory is toward risk surfaces conditioned by both forcing fields and landscape configuration.
  • Transferable Earth-observation and mechanics methods emphasize structured representations and physical constraints: The non-hazard-specific methods cohort repeatedly encodes geometry, spectra, scale, or constitutive structure rather than relying on generic end-to-end automation. GeoVP, spectral dual matching, Mamba-based detection, SPARC-Net, hyperspectral prototype mining, SFMNet, nonstationary random fields, poroelastic wave modeling, and granular-fluid rheology all strengthen analytical capability, but their relevance to landslides or other geohazards remains prospective until domain testing is shown.

Selected Papers

The selected papers combine hazard-specific studies of erosion, landslides, liquefaction, deformation, flooding, wildfire effects, and cascading recovery with a secondary methods cohort in remote sensing, data assimilation, constitutive modeling, and Earth-system datasets. The first group reports substantive geohazard findings; the second offers transferable capability that still requires hazard-specific validation.

1. Integration of RUSLE Modelling and Sentinel-1 and Sentinel-2 Imagery to Assess Bare Ground Evolution Over a Typhoon Season in the Gaoping Basin, Taiwan

Source: Earth Surface Processes and Landforms Type: erosion hazard mapping study Geohazard Type: typhoon-driven soil erosion Relevance: 8/10

Core Problem: Predicting where typhoon rainfall will amplify bare-ground erosion across the Gaoping catchment.

Key Innovation: Integrates RUSLE with Sentinel imagery to track and forecast post-typhoon bare-ground erosion at 10 m resolution.

2. Seismic anisotropy reveals stress rotation caused by magmatic-tectonic interactions in the Natron rift

Source: Frontiers in Earth Science Type: tectonic-volcanic geohazard study Geohazard Type: volcanic and seismic deformation Relevance: 8/10

Core Problem: Determining how magma transport and tectonic extension jointly modify stress and deformation in the Natron Rift.

Key Innovation: Uses dense local shear-wave splitting to resolve lateral, depth-dependent, and temporal stress-field changes tied to magma storage and volcanic activity.

3. SiamNet: A Double-Temporal SAR Avalanche Detection Method Integrating Multiscale Features and an Attention Mechanism

Source: Remote Sensing (MDPI) Type: SAR avalanche detection Geohazard Type: avalanche Relevance: 8/10

Core Problem: Existing SAR avalanche change detection misses small events and needs region-specific tuning.

Key Innovation: Siamese multiscale attention network that improves small-avalanche detection from double-temporal Sentinel-1 SAR.

4. Multiscale Decomposition Reveals the Scale-Dependent Drivers of Complex Urban Ground Deformation in Zhengzhou, China

Source: Remote Sensing (MDPI) Type: urban ground deformation source separation Geohazard Type: ground deformation Relevance: 8/10

Core Problem: Urban deformation patterns blend groundwater, construction, and land-use signals across scales.

Key Innovation: SBAS-InSAR combined with wavelet decomposition and ICA to isolate large-scale and local deformation modes.

5. Evidence for distributed transpressional deformation and seismic hazard implications from the 2025 Mw 6.0 Asadabad- Kunar earthquake, Eastern Afghanistan

Source: Natural Hazards Type: earthquake deformation and hazard implications Geohazard Type: earthquake and coseismic landslides Relevance: 8/10

Core Problem: Mapped fault traces alone may underestimate the width of active deformation and secondary hazards after the 2025 Kunar event.

Key Innovation: Field observations and PSInSAR jointly show distributed transpressional deformation across multiple interacting structures.

6. Hydrological perspective of landslide dams: a review

Source: Natural Hazards Type: landslide dam hydrology review Geohazard Type: landslide dam and flash flood Relevance: 8/10

Core Problem: Hydrologic processes controlling landslide-dam evolution and failure are under-synthesized relative to geomorphic factors.

Key Innovation: Review reframes landslide-dam behavior around inflow, storage, erosion, and remote-sensing and modeling of scarce-data basins.

7. Enhancing the performance of medium-resolution DEM-based landslide susceptibility maps with machine learning algorithms: a case study of Watauga County, North Carolina

Source: Geoenvironmental Disasters Type: landslide susceptibility mapping Geohazard Type: landslide Relevance: 8/10

Core Problem: Landslide susceptibility mapping degrades where only medium-resolution DEMs are available.

Key Innovation: Shows that weighted integration of multiple 30 m DEMs can approach LiDAR-based susceptibility performance with ML.

8. Evidence for Mechanical Interactions Between Precursory Rockfalls, Pre-failure Deformation, and Future Rockfalls

Source: Rock Mechanics and Rock Engineering Type: rockfall precursor process analysis Geohazard Type: rockfall Relevance: 8/10

Core Problem: It is unclear whether clustered precursory rockfalls are mechanically meaningful or just random co-location.

Key Innovation: Null-model simulations show precursory rockfalls are non-random and mechanically linked to future failures.

9. Common and type-specific geo-environmental controls of landslide-type variability: A multivariate analysis in the Rif Mountains, Morocco

Source: Geomorphology Type: landslide type-control analysis Geohazard Type: landslide Relevance: 8/10

Core Problem: Most landslide studies aggregate movement types, obscuring type-specific geo-environmental controls.

Key Innovation: Uses a field-validated inventory with PCA and discriminant analysis to distinguish shared versus type-specific controls on landslide types.

10. Cyclic deformation regimes in liquefiable sands and sheet-pile systems

Source: Soil Dynamics and Earthquake Engineering Type: liquefaction-induced deformation mechanics Geohazard Type: liquefaction Relevance: 8/10

Core Problem: System response in liquefiable deposits depends on transitions between cyclic mobility and residual deformation regimes.

Key Innovation: Validated constitutive and boundary-value framework explaining displacement plateaus in sheet-pile systems.

11. Framework for Cross-Disaster Building Damage Assessment Using Cost-Sensitive Learning

Source: Remote Sensing (MDPI) Type: cross-disaster building damage assessment Geohazard Type: multi-hazard disaster damage Relevance: 7/10

Core Problem: Building damage classifiers struggle with scarce, imbalanced, and cross-event imagery.

Key Innovation: Segmentation-guided cost-sensitive framework evaluated for transfer across optical and SAR disaster datasets.

12. Long-Term InSAR Monitoring and Anomaly Detection of Railway Deformation in Shanghai

Source: Remote Sensing (MDPI) Type: InSAR deformation anomaly detection Geohazard Type: land subsidence Relevance: 7/10

Core Problem: Railway subsidence signals mix broad background settlement with localized engineering anomalies.

Key Innovation: SBAS-InSAR plus RF-SHAP attribution and LightGBM residual analysis to separate mechanisms and flag high-risk anomalies.

13. Urban expansion across a mountain-to-coastal gradient and its implications for hydrogeomorphic risk in Port-au-Prince, Haiti

Source: Natural Hazards Type: urban flood susceptibility and hydrogeomorphic risk Geohazard Type: flood Relevance: 7/10

Core Problem: Rapid urban expansion across slope-to-lowland terrain alters runoff pathways and intensifies flood exposure.

Key Innovation: Mountain-to-coast land-change analysis combined with susceptibility modeling and multi-product extreme-rainfall trends.

14. AIoT-enabled urban platform for flood detection and impact mapping: towards near-real-time spatial decision support in disaster management

Source: Natural Hazards Type: near-real-time flood detection and impact mapping platform Geohazard Type: flood Relevance: 7/10

Core Problem: Smart-city systems often stop at flood extent mapping and do not automate exposure and impact analytics.

Key Innovation: AIoT platform that couples Sentinel-2 flood segmentation with infrastructure and population impact mapping in ArcGIS.

15. Life-Cycle Reliability Evolution and Resilience Enhancement of Anchored Slope Under Stochastic Anchor Performance Degradation

Source: Rock Mechanics and Rock Engineering Type: anchored slope reliability and resilience Geohazard Type: slope instability Relevance: 7/10

Core Problem: Anchor degradation and incomplete inspections complicate long-term reliability management of anchored slopes.

Key Innovation: Life-cycle framework combining censored-data hazard modeling, 3D slope mechanics, and resilience-cost optimization.

16. Recov-Vision: Linking street view imagery and vision-language models for post-disaster recovery

Source: International Journal of Disaster Risk Reduction Type: vision-language post-disaster recovery analysis Geohazard Type: post-hurricane recovery Relevance: 7/10

Core Problem: Overhead imagery misses facade and access cues needed to assess parcel-scale recovery and occupancy after disasters.

Key Innovation: Links panoramic street video to parcels, extracts interpretable facade attributes, and compares transparent and two-stage LLM decision policies for occupancy-recovery assessment.

17. GeoVP: A Unified Visual Prompting Framework for Multisource Remote-Sensing Image Understanding

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: multisource remote-sensing foundation method Geohazard Type: None Relevance: 6/10

Core Problem: Making prompt-based region understanding work reliably across heterogeneous remote-sensing modalities and prompt types.

Key Innovation: Unifies point, box, and free-form prompting for optical, SAR, and infrared imagery with a region-aware multimodal framework and a large remote-sensing prompt dataset.

18. Assimilation of FY-3G Precipitation Data Using a Machine Learning-Based Observation Operator in the CMA-MESO Regional Model

Source: Remote Sensing (MDPI) Type: ML precipitation data assimilation Geohazard Type: extreme rainfall Relevance: 6/10

Core Problem: Conventional precipitation observation operators inject large uncertainty into convective-scale NWP assimilation.

Key Innovation: Machine-learning observation operator embedded in a 1D-Var plus 3D-Var framework for FY-3G precipitation assimilation.

19. A simple constitutive framework for thinning-thickening transitions in dense granular-fluid mixtures

Source: Acta Geotechnica Type: granular-fluid rheology and instability modeling Geohazard Type: None Relevance: 6/10

Core Problem: Dense granular-fluid mixtures can switch between shear thinning and shear thickening in ways standard models miss.

Key Innovation: Unified constitutive framework coupling hypoplastic, suspension-stress, and frictional-contact transition physics.

20. Wildfire Effects on the Geotechnical Properties of Soils: A State-of-the-Art Review

Source: Geotechnical and Geological Engineering Type: post-wildfire soil geotechnics review Geohazard Type: wildfire Relevance: 6/10

Core Problem: Wildfire-driven changes in soil strength, hydraulic behavior, and hydrophobicity remain fragmented across the literature.

Key Innovation: Synthesizes wildfire characterization methods and wildfire-induced changes in key soil geotechnical properties while identifying gaps relevant to subsequent hazards.

21. Wave Propagation in 3-D Poroelastic Media Accounting for Both Mesoscopic and Global Flow Effects

Source: Journal of Geophysical Research: Solid Earth Type: poroelastic wave propagation method Geohazard Type: None Relevance: 5/10

Core Problem: Simulating wave-induced fluid-flow effects in heterogeneous porous media without impractical frequency sweeps or oversized domains.

Key Innovation: Introduces a dimensionless dynamic framework that captures mesoscopic and global flow effects in 3D poroelastic media.

22. Hyperspectral Reconstruction Using Spectral Dual Matching Transformer Through Channel Decoupling and Fusing

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: hyperspectral reconstruction method Geohazard Type: None Relevance: 5/10

Core Problem: Recovering hyperspectral cubes from RGB despite severe spectral overlap and entangled channel responses.

Key Innovation: Decouples RGB channels and combines sparse-dense spectral matching with cross-channel fusion in a transformer framework.

23. Mamba-Enhanced Lightweight Remote Sensing Object Detection

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: remote-sensing object detection method Geohazard Type: None Relevance: 5/10

Core Problem: Balancing lightweight deployment with accurate rotated-object detection in large remote-sensing images.

Key Innovation: Injects Mamba state-space modeling and multiscale receptive aggregation into a compact rotated-object detector.

24. SPARC-Net: Spectral-Preserving Amplitude-Phase and Reliable Correction Network for Long-Tailed Hyperspectral Image Classification

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: hyperspectral classification method Geohazard Type: None Relevance: 5/10

Core Problem: Classifying long-tailed hyperspectral data without losing spectral continuity or over-biasing to head classes.

Key Innovation: Preserves amplitude-phase spectral structure and adds reliability-aware prototype correction with staged training and tail-prior calibration.

25. Hyperspectral Object Detection via Spectral Prototype Mining and Spatial-Spectral Mamba Fusion

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: hyperspectral object detection method Geohazard Type: None Relevance: 5/10

Core Problem: Detecting objects in hyperspectral imagery despite redundant bands and difficult spatial-spectral interactions.

Key Innovation: Pairs spectral prototype mining with spatial-spectral Mamba fusion and adaptive cross-domain feature balancing.

26. Enhancing the Spatial Applicability of SMAP Soil Moisture Using Multi-Stage Machine Learning Downscaling

Source: Remote Sensing (MDPI) Type: soil moisture downscaling Geohazard Type: None Relevance: 5/10

Core Problem: Coarse SMAP soil moisture is hard to use in heterogeneous local terrains.

Key Innovation: Multi-stage Random Forest downscaling and refinement from 9 km to 100 m using retrieval-relevant and regional controls.

27. Measuring heat stress and mitigation capacity around transit stops during Vancouver's 2021 heat dome using hyperlocal microclimate simulations

Source: International Journal of Disaster Risk Reduction Type: urban heat-stress microclimate analysis Geohazard Type: extreme heat Relevance: 5/10

Core Problem: Heat-dome exposure and mitigation capacity around transit stops are hard to measure at hyperlocal scale.

Key Innovation: Uses hyperlocal microclimate simulations to quantify heat stress and mitigation around urban transit stops.

28. Irrigation Impact on Humid Heat Stress: Divergent Metrics, Vegetation Dynamics, and Future Worsening

Source: Geophysical Research Letters Type: climate hazard modeling study Geohazard Type: humid heat stress Relevance: 4/10

Core Problem: Clarifying how irrigation and vegetation dynamics alter humid heat stress metrics under warming.

Key Innovation: Shows that irrigation impacts reverse sign across heat-stress metrics and are amplified when dynamic vegetation is modeled.

29. Spatiotemporal Dynamics and Driving Factors Analysis of Vegetation Net Primary Productivity in China Using Remote Sensing and Machine Learning

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: ecohydrological remote-sensing analysis Geohazard Type: None Relevance: 4/10

Core Problem: Explaining seasonal and interannual NPP dynamics and their drivers across China's major river basins.

Key Innovation: Combines remote sensing, machine learning, SHAP analysis, and spatial autocorrelation modeling to attribute basin-scale NPP change.

30. Orthogonal Complement Subspace Projection for Hybrid Baseline Ocean Current Inversion

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: SAR ocean-current inversion method Geohazard Type: None Relevance: 4/10

Core Problem: Decoupling sea-surface velocity from elevation phase in hybrid-baseline ATI SAR observations.

Key Innovation: Projects multibaseline interferometric observations into the orthogonal complement of the elevation vector to suppress coupling and noise.

31. Two-Stage Framework for Cropland Change Detection in Remote Sensing via Synergistic Feature Enhancement and Temporal Trend Analysis

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Type: remote-sensing change-detection method Geohazard Type: None Relevance: 4/10

Core Problem: Separating real cropland conversion from transient phenological variation in very-high-resolution imagery.

Key Innovation: Decouples spatial feature enhancement from temporal trend analysis using spatial priors plus long-term NDVI mixing.

32. GIFT-BDS: A high-resolution TEC and Gradient Ionospheric Index dataset over China derived from BeiDou GEO fixed-geometry observations

Source: ESSD Type: ionospheric geospatial dataset paper Geohazard Type: None Relevance: 4/10

Core Problem: Resolving fine-scale TEC and ionospheric gradient variability over China more sharply than conventional products.

Key Innovation: Uses fixed-geometry BeiDou GEO observations to derive a high-resolution TEC and gradient index dataset.

33. Updating the ESA Earth System Model for Future Gravity Mission Simulation Studies: ESA ESM 3.0

Source: ESSD Type: Earth system simulation dataset paper Geohazard Type: None Relevance: 4/10

Core Problem: Updating a synthetic Earth system model to test the capabilities of future satellite gravity missions.

Key Innovation: Expands ESA ESM 3.0 with updated atmosphere, ocean, hydrology, solid-Earth, sea-level, and earthquake signal simulations.

34. Mental health consequences of riverbank erosion: examining anxiety and PTSD among affected communities in Southwestern Bangladesh

Source: Natural Hazards Type: riverbank-erosion mental-health impact study Geohazard Type: riverbank erosion Relevance: 4/10

Core Problem: The anxiety and post-traumatic stress burden associated with recurrent riverbank erosion and displacement remains poorly quantified.

Key Innovation: Compares erosion-exposed and unexposed residents and identifies displacement, livestock loss, and social support as determinants of anxiety and PTSD.

35. Solidified dredged soil for offshore wind turbine scour protection: mechanical properties, microscopic mechanisms, and carbon emission assessment

Source: Acta Geotechnica Type: local-scour protection material study Geohazard Type: local scour Relevance: 4/10

Core Problem: Offshore foundations need pumpable scour-protection material that resists underwater dispersion while limiting cost and carbon emissions.

Key Innovation: Develops dredged-soil-based fluidized solidified soil with quantified flowability, strength, dispersion resistance, microstructure, cost, and carbon performance.

36. Effect of water-retaining agent on MICP solidification of aeolian sand

Source: Acta Geotechnica Type: aeolian sand stabilization Geohazard Type: wind erosion Relevance: 4/10

Core Problem: Rapid evaporation limits MICP effectiveness for stabilizing desert sand against wind erosion.

Key Innovation: Water-retention-assisted MICP that sustains microbial activity and improves sand crust strength and erosion resistance.

37. Unified characterization of soil heterogeneity using nonstationary random fields

Source: Engineering Geology Type: soil heterogeneity uncertainty modeling Geohazard Type: None Relevance: 4/10

Core Problem: Fusing multi-source subsurface data into a unified nonstationary representation of soil heterogeneity.

Key Innovation: Nonstationary random-field framework for unified characterization of subsurface heterogeneity.

38. A scientometric analysis of digital twin technology in flood disaster management with practical implementation insights

Source: International Journal of Disaster Risk Reduction Type: scientometric flood digital-twin review Geohazard Type: flood Relevance: 4/10

Core Problem: Flood-management applications of digital twins are expanding, but the research landscape and practical implementation pathways remain insufficiently synthesized.

Key Innovation: Combines scientometric mapping of flood digital-twin research with real-world case synthesis and a practical coastal implementation framework.

39. Post-earthquake fire preparedness and operational challenges in Portuguese fire services: Evidence from a questionnaire-based survey

Source: International Journal of Disaster Risk Reduction Type: post-earthquake fire preparedness survey Geohazard Type: post-earthquake fire Relevance: 4/10

Core Problem: Preparedness gaps in Portuguese fire services for post-earthquake fire scenarios are poorly documented.

Key Innovation: First focused questionnaire-based assessment of Portuguese fire-service vulnerabilities under PEF conditions.

40. SFMNet: Joint semantic and structural modeling via spatial-frequency collaboration for remote sensing scene segmentation

Source: International Journal of Applied Earth Observation and Geoinformation Type: remote sensing scene segmentation method Geohazard Type: None Relevance: 4/10

Core Problem: Remote-sensing scene segmentation needs stronger joint modeling of semantic context and structural detail.

Key Innovation: Spatial-frequency collaborative network for jointly modeling semantic and structural cues in scene segmentation.

41. The influence of hydrophobic agent on unsaturated seepage mechanisms in silty clay: insights from electrical resistivity monitoring

Source: Journal of Hydrology Type: unsaturated seepage monitoring method Geohazard Type: None Relevance: 4/10

Core Problem: Moisture-dependent hydrophobicity makes transient infiltration and barrier performance difficult to characterize with equilibrium methods.

Key Innovation: Combines multilevel electrical-resistivity monitoring with the transient-profile method and a content-dependent Gardner model to estimate water migration and unsaturated hydraulic conductivity continuously.