TerraMosaic Daily Digest: Mar 13, 2026
Daily Summary
Research associated with the March 13, 2026 digest converges on a more explicit treatment of hazard cascades. The strongest papers do not stop at identifying a slope failure, a debris flow, or a dammed channel in isolation. They reconstruct how extreme rainfall mobilized a sequence of landslides, debris flows, hyperconcentrated flows, and channel widening in Himachal Pradesh; how a recurrent landslide dam in Taiwan formed, failed, and re-formed under repeated monitoring; and how debris-flow peak discharge and volume can now be measured directly with field-resolved 3D LiDAR rather than inferred from bulk assumptions. Across these studies, the scientific gain lies in resolving transfer processes between hillslopes, channels, and infrastructure.
The second axis of progress is state reconstruction under incomplete observations. Flooding is detected with privacy-preserving on-device cameras, hydrogeological events are assembled into national-scale early warning records, and coseismic deformation is retrieved directly from single interferograms. Cold-region work is equally process-based: freeze-thaw sorting on talus slopes, ventilation-driven permafrost protection, brittle adfreeze shear, and tunnel response in frozen saturated ground are all handled as coupled thermal-hydraulic-mechanical systems. Even the hydrological papers are less about generic forecasting than about memory, storage, and spatial coherence, whether in long-duration drought, ski-slope runoff disturbance, or distributed erosion prioritization.
Key Trends
Today's literature is marked by slope-channel cascade analysis, observation-driven hazard measurement, and infrastructure models that now combine structural, hydrological, and functional states.
- Rainfall disasters are being resolved as coupled geomorphic cascades: the most informative studies connect hillslope release, sediment transformation, channel adjustment, and built-environment consequence within a single event logic.
- Hazard observation is becoming spatially explicit and operational: drone and satellite repeat surveys, 3D LiDAR, multisource disaster databases, and on-device machine learning are reducing dependence on sparse proxy measurements.
- Cold-region geohazards are now treated as fully coupled state problems: freeze-thaw sorting, permafrost railbed cooling, adfreeze shear, and frozen-ground tunnel response all depend on thermal, hydraulic, and mechanical co-evolution.
- Infrastructure resilience studies are moving beyond single failure criteria: tunnel serviceability, bridge restraint systems, embankment seepage, and fault reactivation are being framed through interacting structural and operational modes.
- Hydrological hazard modelling is increasingly memory-aware: drought and runoff papers show that distributed storage, long-term basin state, and land-use disturbance matter as much as short-term forcing.
Selected Papers
This digest features 22 selected papers from 636 papers analyzed across landslide and debris-flow cascades, landslide-dam monitoring, seismic slope failure, coastal and drought hazards, cold-region infrastructure, and operational geohazard sensing.
1. Cascading rainfall-induced sediment disaster in Mandi District, Himachal Pradesh (2025): multi-hazard characterization and microstructural insights
Core Problem: Extreme-rainfall disasters in steep Himalayan terrain are still too often parsed as separate failures rather than coupled slope-channel cascades.
Key Innovation: Field mapping, mineralogical analysis, SEM/EDS, and rheology are combined to show how fine-rich, shear-thinning debris amplified landslide-to-channel cascade behavior in the 2025 Mandi disaster.
2. Monitoring and characteristics of formation and failure of a recurrent landslide dam: A case study of the Danan River landslide dam in Taitung, Taiwan
Core Problem: Recurrent landslide dams in remote mountain terrain are difficult to survey and reassess as their stability evolves through repeated collapse.
Key Innovation: The paper proposes a remote stability-review workflow combining satellite images, helicopter surveys, drone photography, and downstream water-level monitoring to document repeated dam formation and failure.
3. How well can we measure peak discharge and volume? Instantaneous 3D LiDAR measurements of multiple debris flows at three locations along a channel
Core Problem: Peak discharge and event volume of debris flows remain biased when estimated from coarse velocity assumptions and cross sections contaminated by deposits.
Key Innovation: Instantaneous 3D LiDAR velocity-depth fields support a column-wise discharge method that resolves event-type-specific biases and improves debris-flow volume and peak-flow estimation.
4. System probability analysis approach for the stability of rock slopes subjected to seismic excitation
Core Problem: Seismic rock-slope assessments usually isolate single sliding planes and therefore miss integrated system instability across multiple failure modes.
Key Innovation: A pseudo-dynamic probabilistic framework is extended to multi-plane series systems with Cornell bounds, enabling system-level instability estimates for seismically loaded rock slopes.
5. A Multisource Information System in support of Early Warning Systems in Brazil: The REINDESC database
Core Problem: Low-severity but high-frequency hydrogeological events remain underreported, weakening threshold-setting and warning across national disaster systems.
Key Innovation: REINDESC builds a multiscale, multisource disaster database for Brazil that captures low-magnitude floods and landslides usually absent from official warning records.
6. Detection of Coastal Flooding With TinyCamML: A Low-Cost, Privacy-Preserving Cellular-Connected Camera With Onboard ML
Core Problem: Hyper-local roadway flooding is difficult to monitor continuously because point sensors miss extent and full-image transmission raises privacy and cost concerns.
Key Innovation: TinyCamML performs on-device flood classification with a low-cost solar-powered camera, transmitting only flood/no-flood decisions while preserving near-real-time spatial awareness.
7. Inverse Grading Emerges From Particle-Scale Migration Under Seasonal Freeze-Thaw Forcing: Evidence From Multi-Year Monitoring and Physical Modeling
Core Problem: Inverse grading on talus slopes is widespread, yet its formation under realistic seasonal forcing remains poorly quantified.
Key Innovation: Multi-year monitoring and freeze-thaw experiments show how finer particles preferentially migrate through expanded pores, producing inverse grading and depth-dependent slope creep.
8. Cooling characteristics of the dual-layer ventilation slope and underlying permafrost response process of the railbed
Core Problem: Climate-sensitive railbeds in permafrost terrain need more effective active cooling strategies than current approaches provide.
Key Innovation: Two-year monitoring and modelling show that a dual-layer ventilation slope substantially enhances convective cooling and can maintain underlying permafrost over multi-decadal timescales.
9. Failure characteristics and development mechanism of time-delayed rockburst in a tunnel excavated via the drilling and blasting method: A case study
Core Problem: Time-delayed rockburst is difficult to anticipate because triggering depends on cumulative stress evolution rather than immediate excavation response.
Key Innovation: Field evidence, triaxial tests, simulation, and microseismic monitoring are integrated to show how stress concentration, time-dependent weakening, and repeated disturbances converged to trigger an intense delayed rockburst.
10. Probabilistic dual-dimensional seismic performance assessment of large-diameter shield tunnels considering lining-internal structure interaction
Core Problem: Tunnel earthquake assessment often ignores how structural safety and internal functionality degrade together.
Key Innovation: A dual-dimensional probabilistic framework combines lining and internal-structure damage into joint performance states and fragility curves for shield tunnels.
11. Gaussian random field-based downscaling for drought projections: linking CMIP6 SSPs to local hydrology in a subtropical river basin
Core Problem: Basin-scale drought projections remain weak when downscaling fails to preserve spatial correlation in climate forcing.
Key Innovation: A Gaussian-random-field framework preserves spatial coherence across CMIP6 scenarios and improves basin-scale drought outlooks when coupled to distributed hydrological modelling.
12. Holistic Retrieval of Absolute Coseismic Displacement Fields From Single Interferograms via Physics-Aware GANs
Core Problem: Conventional coseismic InSAR processing remains too sequential and fragile for fully automated rapid-response mapping.
Key Innovation: A physics-aware GAN retrieves absolute coseismic displacement directly from single noisy interferograms, eliminating much of the usual post-processing chain.
13. Can Conceptual Rainfall-Runoff Models Capture Multi-Annual Storage Dynamics?
Core Problem: Many conceptual rainfall-runoff models fail during prolonged drought because they do not capture slow storage depletion and delayed recovery.
Key Innovation: The study identifies the minimum model structure required to represent multi-annual storage dynamics: a dedicated slow store, disconnection from direct streamflow, and gradual long-term water loss.
14. Seepage-erosion induced failure analysis of poorly graded sand using transparent soil visualization technique
Core Problem: Internal erosion and seepage failure remain difficult to observe directly, limiting mechanistic understanding of unstable sandy skeleton collapse.
Key Innovation: Transparent-soil visualization and PIV reveal dominant flow zones, particle transport pathways, and the role of fine content during seepage-induced failure of poorly graded sand.
15. Experimental study and interpretation of internal and adfreeze shear behavior of frozen clay till and at clay till-steel pile interface
Core Problem: Pile uplift risk in freezing ground cannot be assessed reliably without better constraints on internal frozen-soil shear and brittle adfreeze bond failure.
Key Innovation: Direct shear experiments isolate how temperature, water content, and shear rate control peak and residual internal and adfreeze shear behavior at the clay till-steel pile interface.
16. Fault reactivation in low-porosity carbonates: The role of fault roughness and confining pressure
Core Problem: The transition between stable sliding and stick-slip reactivation in carbonate faults remains poorly linked to geometry and confinement.
Key Innovation: Triaxial shear tests show how roughness and confining pressure partition strain between matrix damage and fault slip, shifting the system from clustered microseismicity toward stable sliding.
17. Active fault characterization: Improving fault scaling laws by drone technology and advanced statistics
Core Problem: Global fault scaling laws remain limited by poor 3D exposure and inconsistent geometric measurement.
Key Innovation: Drone photogrammetry and refined statistics update active-fault scaling laws and clarify where maximum geometric attributes localize within fault zones.
18. Hydrological drivers of surface runoff during high intensity rainfall experiments in Alpine ski regions
Core Problem: Ski-slope construction alters runoff behavior, but the dominant controls on rapid storm runoff in Alpine resorts remain insufficiently resolved.
Key Innovation: Rainfall-simulation experiments across 12 ski regions show that disturbed slope preparation, geology, and hydrological state together produce much higher runoff coefficients on ski slopes.
19. Integrated morphometric and LULC change-based soil erosion risk prioritization in the Gaudi River sub-watersheds, Lesser Himalaya
Core Problem: Watershed-scale erosion control remains inefficient when terrain susceptibility and land-cover degradation are evaluated separately.
Key Innovation: Morphometric indices and multi-decadal land-cover change are integrated to prioritize erosion-prone sub-watersheds in the Lesser Himalaya for targeted intervention.
20. Seismic response of a composite-lined tunnel in saturated frozen soil half-space to plane P1-wave
Core Problem: Tunnel design in frozen saturated ground lacks systematic understanding of how temperature, burial depth, and lining configuration reshape seismic demand.
Key Innovation: A Fourier-Bessel solution framework resolves displacement and stress concentration in composite-lined tunnels under P-wave loading across cold-region design parameters.
21. Seismic performance of longitudinal shear keys in preventing girder unseating in highway bridges
Core Problem: Bridge girder-unseating mitigation requires restraint parameters that reduce unseating without transferring unacceptable damage elsewhere.
Key Innovation: Nonlinear analyses and fragility assessment identify shear-key capacity and gap ranges that suppress girder unseating while limiting substructure damage.
22. A physically consistent soil thickness map of the Qinghai-Tibet Plateau derived from coupled erosion mechanisms
Core Problem: Physically realistic soil-thickness data remain sparse across the Qinghai-Tibet Plateau, limiting distributed hydrological, ecological, and slope-process modelling.
Key Innovation: A mechanistic mass-balance approach coupling weathering with wind, hydraulic, and gravitational erosion produces a plateau-scale soil-thickness dataset that outperforms purely empirical maps.