TerraMosaic Daily Digest: Jan 23, 2026
Daily Summary
Today's landslide research spans fundamental geomaterial properties, advanced remote sensing, and hazard/risk assessment frameworks. Across the new papers, a major theme is improving landslide prediction and susceptibility mapping by combining machine learning with multi-sensor remote sensing and data-fusion workflows.
Another strong cluster focuses on climate-sensitive slopes—especially thawing permafrost terrain and landscapes affected by wildfire—and on hydrological triggers such as rainfall infiltration, groundwater dynamics, and river regulation. Several studies translate these insights into engineering practice for infrastructure (e.g., tunnels, roads, and railways), aiming for more reliable hazard assessment and decision support.
Key Trends
- AI/Remote Sensing Integration: A surge in studies leveraging machine learning and AI for geohazard susceptibility mapping, change detection, and image registration, often incorporating multi-source remote sensing data.
- Permafrost and Climate Change Impacts: Significant research focused on understanding the effects of thawing permafrost and changing climate conditions on slope stability, particularly in regions like the Tibetan Plateau and high-latitude areas.
- Hydrological Influences: A continued emphasis on the role of water, including rainfall infiltration, groundwater dynamics, and river regulation, in triggering and influencing landslide behavior.
Selected Papers
This digest features 70 selected papers from 2,070 papers analyzed across multiple journals (duplicates in the source list removed during page generation). Each paper has been evaluated for its relevance to landslide research and includes links to the original publications.
1. Spring trends in slow-moving landslide displacement: is it a reliable way to predict their movements? Two case studies in the Susa Valley (NW Italy)
Core Problem: Predicting acceleration patterns in slow-moving landslides is complicated by multiple interacting factors.
Key Innovation: Using spring water levels as a proxy for predicting slope displacements and comparing this approach with piezometric data.
2. Effect of depositional angle on the water-holding characteristics of transversely isotropic unsaturated loess
Core Problem: Water-holding characteristics of loess affect the stability and safety of engineering structures.
Key Innovation: Investigating how different depositional angles influence the water-holding characteristics of transversely isotropic unsaturated loess.
3. Experimental study on mechanical properties of frozen silty clay containing ice lenses
Core Problem: Ice lenses in permafrost slopes affect mechanical properties.
Key Innovation: Triaxial tests on frozen silty clay with ice lenses under different temperatures, inclinations, thicknesses, and confining pressures.
4. Enhancing geohazard susceptibility mapping with a sample reliability algorithm: a case study in Guangzhou, China
Core Problem: Machine learning performance in geohazard susceptibility assessment is constrained by the scarcity and uncertain reliability of geohazard samples.
Key Innovation: Introducing a Sample Enhancement Algorithm (SEA) to improve training data quality by quantifying sample reliability through a weighted environmental similarity approach. Editor's Choice: tackles uncertain/limited inventories by improving training-sample reliability (via environmental similarity), strengthening ML-based susceptibility mapping.
5. Landslide displacement prediction based on deep displacement state recognition and similarity propagation
Core Problem: Current landslide forecasting is constrained by an over-reliance on surface kinematics, often failing to capture the hidden, non-linear state transitions occurring at depth.
Key Innovation: A Trend–State Coupled Framework that shifts the paradigm from purely external-driven to internal-state-aware prediction.
6. Lithology dependent pathways of forming rock mass weakness plane shape regional landslide susceptibility
Core Problem: Rock mass weakness planes are mechanically unfavored discontinuities that control slope stability, but their spatial pattern is poorly understood.
Key Innovation: A multi-explainable machine learning framework to predict the distribution of weakness planes and quantify their contribution to landslides.
7. Evolution of rockfall risk following changes in hazard and exposure: Application to a road section in the Zermatt valley, Swiss Alps
Core Problem: Rockfall risk assessment needs to account for non-stationary conditions due to socio-environmental transformations, including climate change and changes in exposure (e.g., road traffic).
Key Innovation: A quantitative risk assessment (QRA) approach that evaluates the respective and overall effects of rockfall frequency and traffic density changes on rockfall risks at three time steps (1960, 2010, and 2060).
8. Switch on tunnel vision: Portable wind tunnels to understand and quantify aeolian processes
Core Problem: Understanding and quantifying wind erosion and dust emissions from soil surfaces in their natural state.
Key Innovation: Using portable wind tunnels to examine soil surfaces independently from natural wind events, allowing for controlled studies of wind-induced entrainment, transport, and redistribution of particles.
9. Centimeter-resolution 4D dynamics of retrogressive thaw slumps from repeat UAV photogrammetry on the Tibetan Plateau
Core Problem: Limited understanding of retrogressive thaw slump (RTS) evolutionary processes due to a lack of high-resolution observations.
Key Innovation: Using repeated centimeter-scale UAV surveys to track the 4D dynamics of RTSs on the Tibetan Plateau, quantifying changes in area, headwall retreat, surface movement, elevation, and volume.
10. A review of forward modelling and retrieval approaches for forest soil moisture and vegetation optical depth using L-band radiometry
Core Problem: Existing algorithms for retrieving soil moisture and vegetation optical depth using L-band radiometry are primarily developed for low-biomass vegetation types and are not suitable for forests.
Key Innovation: Reviewing current retrieval approaches, performance assessment methods, and available validation resources to identify the limitations and challenges in forest retrievals using L-band radiometry.
11. High-resolution annual desertification mapping in northern China using a novel comprehensive desertification index and unsupervised algorithm
Core Problem: The lack of high-resolution desertification products and unclear remote sensing mechanisms hinder accurate desertification monitoring.
Key Innovation: Constructing a comprehensive desertification index (CDI) integrating multisource remote sensing data and applying a Gaussian mixture model (GMM) for automated 10m-resolution annual desertification mapping.
12. Temporal attention multi-resolution fusion of satellite image time-series, applied to Landsat-8/9 and Sentinel-2: all bands, any time, at best spatial resolution
Core Problem: Fusing Satellite Image Time Series (SITS) data from multiple sensors with variable spatial resolutions and acquisition times is challenging.
Key Innovation: A novel Masked Auto-Encoder training strategy and deep learning architecture (TAMRF-SITS) are proposed to fuse SITS data from multiple sensors, predicting all spectral bands at the best spatial resolution and any acquisition time.
13. Began+: Leveraging bi-temporal SAR-optical data fusion to reconstruct clear-sky satellite imagery under large cloud cover
Core Problem: Reconstructing clear-sky satellite imagery from cloudy images, especially with large cloud cover, is challenging due to insufficient attention to temporal changes and limitations of deep models.
Key Innovation: A novel deep learning framework, Began+, integrates bi-temporal SAR-optical data with an enhanced generative adversarial network and post-processing to reconstruct clear-sky imagery under high cloud cover, capturing bi-temporal change features.
14. Annotation-free cloud masking for PlanetScope images in the Arctic via cross-platform ability transfer using deep learning and foundation models
Core Problem: Cloud masking in the Arctic is challenging due to the similarity between clouds and snow/ice, and the lack of annotated data for PlanetScope images.
Key Innovation: A cross-platform ability transfer paradigm is proposed, using a new dataset of coincident PlanetScope and Landsat-8 image pairs to transfer cloud masking skills from Landsat-8 to PlanetScope without manual annotations.
15. PANet: A multi-scale temporal decoupling network and its high-resolution benchmark dataset for detecting pseudo changes in cropland non-agriculturalization
Core Problem: Detecting cropland non-agriculturalization (CNA) is difficult due to spectral confusion from seasonal fluctuations, weather interference, and imaging discrepancies.
Key Innovation: A phenology-aware temporal change detection network (PANet) is proposed, leveraging a dual-driven decoupling model and a sample balance adjustment module to improve CNA detection accuracy.
16. Weak supervision makes strong details: fine-grained object recognition in remote sensing images via regional diffusion with VLM
Core Problem: Fine-grained object recognition (FGOR) in remote sensing images (RSIs) is constrained by reliance on labeled data, difficulty reconstructing details in low-resolution images, and limited robustness.
Key Innovation: An automatic FGOR network (AutoFGOR) is proposed, using a hierarchical dual-pipeline architecture with regional diffusion and a vision language model (VLM) for object region super-resolution reconstruction.
17. A weakly supervised approach for large-scale agricultural parcel extraction from VHR imagery via foundation models and adaptive noise correction
Core Problem: Extracting agricultural parcels from very-high-resolution (VHR) imagery is challenging due to costly manual annotations and the limited capacity of segmentation architectures.
Key Innovation: A Weakly Supervised approach for agricultural Parcel Extraction (WSPE) is proposed, integrating tabular and vision foundation models for pseudo-label generation and adaptive noisy label correction.
18. Progressive uncertainty-guided network for binary segmentation in high-resolution remote sensing imagery
Core Problem: Binary semantic segmentation in remote sensing (RS) imagery faces challenges due to complex object appearances, ambiguous boundaries, and high similarity between foreground and background.
Key Innovation: The Progressive Uncertainty-Guided Segmentation Network (PUGNet) is proposed, explicitly modeling uncertainty in a context-aware manner with specialized modules for foreground, background, and contextual uncertainty.
19. Mamba-CNN hybrid Multi-scale ship detection Network driven by a Dual-perception feature of Doppler and Scattering
Core Problem: Ship detection in polarimetric SAR (PolSAR) imagery faces challenges in discriminating targets from clutter, detecting multi-scale objects, and achieving real-time detection.
Key Innovation: A Mamba-CNN hybrid Multi-scale ship detection Network (MCMN) is proposed, driven by a Dual-perception feature of Doppler and Scattering (DDS) and using a Multi-scale Information Perception Module (MIPM) and a Local-Global Feature Enhancement Module (LGFEM).
20. SmartQSM: a novel quantitative structure model using sparse-convolution-based point cloud contraction for reconstruction and analysis of individual tree architecture
Core Problem: Tree architecture analysis is challenging due to the complexity of trees, which affects the accuracy and efficiency of point-cloud-based reconstruction.
Key Innovation: SmartQSM, a novel quantitative structure model, is proposed for reconstructing individual trees using ground-based laser scanning data, achieving point cloud contraction with a sparse-convolution-based residual U-shaped network (ResUNet).
21. Unveiling spatiotemporal forest cover patterns breaking the cloud barrier: Annual 30 m mapping in cloud-prone southern China from 2000 to 2020
Core Problem: Monitoring forest cover in persistently cloudy regions like southern China is difficult due to the scarcity of high-quality remote sensing data and reliable training samples.
Key Innovation: A novel forest–non-forest mapping framework is proposed, based on reconstructed remote sensing data using deep learning-based multi-sensor fusion and a spectrally similar sample transfer method.
22. MARSNet: A Mamba-driven adaptive framework for robust multisource remote sensing image matching in noisy environments
Core Problem: Semi-dense matching of multi-source remote sensing images under noise interference is challenging due to low efficiency and reduced performance with large viewpoint variations.
Key Innovation: A hybrid network for multi-source remote sensing image matching based on an efficient and robust Mamba framework, named MARSNet, is proposed, leveraging the Mamba network and a frozen pre-trained DINOv2 foundation model.
23. RSMT: Robust stereo matching training with geometric correction, clean pixel selection and loss weighting
Core Problem: Inaccurate or noisy labels have a huge impact on the training of deep learning models for satellite image stereo matching.
Key Innovation: A robust stereo matching training framework (RSMT) is proposed, with geometric correction, clean pixel selection, and loss weighting modules to deal with label errors at the training level.
24. SuperMapNet for long-range and high-accuracy vectorized HD map construction
Core Problem: Vectorized high-definition (HD) map construction suffers from limited perception capability and range, and low accuracy due to neglecting information between elements.
Key Innovation: SuperMapNet, a multi-modal framework, is proposed for long-range and high-accuracy vectorized HD map construction, using camera images and LiDAR point clouds with tight coupling of semantic and geometric information.
25. SAR-NanoShipNet: A scale-adaptive network for robust small ship detection in SAR imagery
Core Problem: Small ship target detection in synthetic aperture radar (SAR) imagery faces challenges such as high speckle noise, difficulty in extracting small target features, and geometric distortion.
Key Innovation: SAR-NanoShipNet is proposed, employing a specialized convolution (DABConv), deformable convolutions, boundary attention mechanisms, and a VerticalCompSPPF module (VC-SPPF) to enhance small ship detection.
26. Towards High spatial resolution and fine-grained fidelity depth reconstruction of single-photon LiDAR with context-aware spatiotemporal modeling
Core Problem: High spatial resolution (SR) data processing in single-photon LiDAR is challenging, and existing deep learning frameworks fail to overcome this challenge.
Key Innovation: A U-Net++ backbone with dense skip connections is adopted, integrating attention-driven modulation and convolution, a 3D triple local-attention fusion module (3D-TriLAF), and an opposite continuous dilation spatial–temporal convolution module (OCDSConv).
27. Mapping land uses following tropical deforestation with location-aware deep learning
Core Problem: Recognizing individual commodities after deforestation is challenging due to spectral similarities, the limited spatial resolution of free satellite imagery, and limited labeled data.
Key Innovation: A deep learning, multi-modal approach is proposed for the recognition of post-deforestation land uses from a time series of Sentinel-2 images, geographic coordinates, and country-level statistics, using a Transformer-based model.
28. A spectral index using generic global endmembers from Landsat multispectral data for mapping urban areas
Core Problem: Quantifying urban land cover from space is challenging due to the diversity and spectral heterogeneity of urban reflectance.
Key Innovation: A spectral index using Landsat global SVD endmembers, termed the Urban Index using Global Endmembers (GEUI), is proposed to highlight and map urban land.
29. Empowering tree-scale monitoring over large areas: Individual tree delineation from high-resolution imagery
Core Problem: Accurate individual tree delineation (ITD) is essential for forest monitoring, but challenges persist, especially in complex forest environments.
Key Innovation: This study evaluates state-of-the-art instance segmentation approaches for ITD, using the largest forest instance-segmentation imagery dataset to date and standardized evaluation protocols.
30. Varying sensitivities of RED-NIR-based vegetation indices to the input reflectance affect the detected long-term trends
Core Problem: Long-term vegetation trends derived from vegetation indices (VIs) can be affected by the inherent differences between VIs calculated from the same input reflectance.
Key Innovation: This study compares long-term trends in six RED-NIR-based VIs, demonstrating that differences in their sensitivities to RED and NIR reflectance lead to inconsistent trends.
31. Effects of wildfire on sandstone outcrops and environmental consequences, Bohemian Switzerland NP, Czech Republic
Core Problem: Wildfires impact on sandstone outcrops is poorly studied. This research documents fire-induced changes in sandstone outcrops.
Key Innovation: Combined time-lapse photo documentation, mineralogical and petrographic studies, in situ moisture and tensile strength measurements, heat transfer modeling, and thermal desorption of mercury to document changes.
32. Impact of dams on river regime and extreme flow events in MIÑO–SIL river basin (NW of the IBERIAN peninsula)
Core Problem: River regulation significantly alters hydrological regimes, but the long-term effects relative to climate and land-use variability are not well understood.
Key Innovation: Evaluates the roles of precipitation, land-use, and dam regulation on river flow regimes and extreme flood events using long-term hydro-meteorological records.
33. Effects of fire severity on soil organic matter: a multi-isotope (C, N, H, O) comparison of wildfires and experimental burns
Core Problem: Accurately assessing fire severity and its impact on soil organic matter (SOM) composition is challenging but crucial for understanding biogeochemical cycles and post-fire recovery.
Key Innovation: Employs a multi-isotope and elemental approach (C, N, H, O) to assess fire-induced changes in SOM quantity and quality across soil burn severity levels in wildfires and controlled experimental burns.
34. Automated night-time fog detection and masking using machine learning from near real-time satellite observations
Core Problem: Fog significantly reduces visibility, impacting transportation and safety, especially in United Arab Emirates (UAE) during winter months. This study develops a machine learning (ML) approach for automated fog detection and masking from near real-time SEVIRI Satellite observations.
Key Innovation: We evaluate six ML models across four training strategies: (1) supervised training using SEVIRI nighttime microphysics Red-Green-Blue (RGB) pixels with Meteorological Aerodrome Reports (METAR) station labels; (2) as (1) but adding the three infrared channels; (3) k-means labels derived from Night Microphysics RGB; and (4) a fusion of station-labeled and k-means-labeled data.
35. Research on the effects of train-induced wind on the thermal environment of tunnels in seasonally frozen regions
Core Problem: The aerodynamic flow induced by high-speed trains during tunnel traversal impacts the thermal environment of cold-region tunnels, affecting their structural integrity.
Key Innovation: Defines two thermal conditions (positive and negative effect) and analyzes the regulatory roles of blocking ratio, train speed, and train length on the tunnel thermal environment using a validated numerical model.
36. Study on the influence of temperature field during thawing and sinking process of tropical undersea tunnel based on pipe curtain freezing method
Core Problem: Tropical undersea tunnels face risks related to thawing and sinking of soft strata, making the study of the thawing temperature field a critical issue.
Key Innovation: Employs physical similarity tests and numerical simulations to elucidate the evolution of the forced thawing temperature field and the thawing behavior of permafrost using the pipe curtain freezing method.
37. Hydro-thermo-mechanical coupling analysis of freeze-thaw process and optimization of freezing scheme in soft clay stratum
Core Problem: Frost heave and thaw settlement of soft clay stratum during Artificial Ground Freezing (AGF) projects.
Key Innovation: Combines laboratory tests, theoretical modeling, and numerical simulations to address the challenges of frost heave and thaw settlement, and optimizes freezing schemes for AGF projects.
38. Research on transmission line icing classification and recognition algorithm based on BiTex-ResNet34
Core Problem: Accurate identification of transmission line icing types is challenging due to limited samples, high similarity between types, and a lack of non-contact methods.
Key Innovation: A BiTex-ResNet34 algorithm is proposed, using a dual-branch architecture to extract raw and texture features, and a Second-order Feature Fusion Module (SK-FM) to improve the model's ability to distinguish between icing types.
39. Study on elevated temperature effect and damage mechanism of frozen soil under impact loading
Core Problem: Analyzing dynamic damage progression in frozen soil is limited by experimental challenges in obtaining high-resolution thermal measurements during transient impact loading.
Key Innovation: An experimental methodology was developed to synchronously measure mechanical properties and temperature variation of frozen soil under impact loading, using a modified split Hopkinson pressure bar (SHPB) coupled with an Infrared Temperature Measurement System (ITMS).
40. Effects of moss cover patterns on hydrodynamic parameters and particle size selectivity during karst erosion under rock surface flow
Core Problem: Karst rocky desertification is exacerbated by soil erosion. Epilithic mosses can regulate surface flow, but their combined effect with slope moss on soil erosion is unclear.
Key Innovation: Quantified the joint effects of moss cover on rock and slope surfaces on erosion, revealing a dual role of epilithic mosses in erosion regulation, either reducing or exacerbating erosion depending on rainfall intensity.
41. A random simplified analysis method to evaluate braced excavation-induced wall deflection considering spatial variability of soil
Core Problem: Wall deflection in braced excavations is significantly influenced by the spatial variability of soil parameters, limiting the applicability of current simplified methods.
Key Innovation: Proposed a random simplified analysis method using an improved beam-spring model and sparse polynomial chaos expansion to efficiently evaluate excavation-induced wall deflection in spatially variable soil.
42. Quantifying the scale effects on shallow foundation bearing capacity induced by shear band on sensitive marine clays
Core Problem: Mapping from small-scale model tests to prototype design remains challenging due to scale effects induced by progressive failure in sensitive marine clays.
Key Innovation: Investigated scale effects using a nonlocal RITSS method with a strain-softening model, proposing novel rescaling coefficients incorporating internal length and clay ductility to improve bearing capacity prediction.
43. Uplift resistance mechanism of pipes in lightweight backfill material of ceramsite
Core Problem: Buried pipelines crossing active fault zones are vulnerable to seismic displacement. Lightweight backfill materials offer a potential mitigation strategy, but their behavior is not well understood.
Key Innovation: Investigated the suitability of ceramsite as a lightweight backfill, combining element tests, model experiments, and DEM simulations to reveal its lower uplift resistance due to lower self-weight and convergent slip surfaces.
44. A hybrid finite-element material-point approach for anchored tunnels in large deformation
Core Problem: The stability of deep-buried tunnels undergoing large deformation depends on the interaction between rock masses and bolt reinforcement systems, which is difficult to simulate accurately.
Key Innovation: Proposed a Hybrid Finite-Element Material-Point (HFEMP) method, modeling rock with MPM and bolts with FEM, validated against physical tests, and developed optimized bolt design charts for tunnel reinforcement.
45. Multi-objective reliability-based design optimization of piles considering soil-structure interaction and soil spatial variability
Core Problem: Pile design optimization often overlooks soil uncertainties and faces challenges in exploring the entire design space due to computational complexity.
Key Innovation: Introduced a multi-objective reliability-based design optimization (MO-RBDO) framework for pile foundations, incorporating soil spatial variability into a calibrated simplified model for efficient surrogate modeling.
46. Improved p-y curve modeling for large-diameter monopiles in asymmetric scour conditions in sandy soils
Core Problem: Scour affects the bearing capacity of monopile foundations for offshore wind turbines, and the geometric asymmetry of scour holes is often not considered.
Key Innovation: Investigated the effects of scour hole asymmetry on monopile bearing capacity using FEM, extracted p-y curves with a Python-API, and improved the existing p-y curve model based on parametric analyses.
47. Assessment of pavement–subgrade deformation in permafrost highways using UAV photogrammetry and ground-penetrating radar: Case study of Qinghai–Tibet highway
Core Problem: Permafrost-related deformation of highway embankments is a major constraint on the long-term serviceability. Freeze–thaw cycles, water migration and heavy traffic loads produce rutting, corrugation and differential settlement at the surface, but their relationship to subsurface anomalies is not yet fully understood.
Key Innovation: Integrates UAV photogrammetry with ground-penetrating radar (GPR) to examine coupled pavement–subgrade behaviour. UAV-derived digital surface models are used to quantify rut depth, roughness and longitudinal/transverse elevation differentials, whereas 2D GPR profiles and depth-dependent reflection-intensity maps are interpreted to identify stratigraphic undulations, localised loosening and the position of the permafrost table.
48. Water migration in frozen high-speed railway subgrades under traffic vibration: Piston suction versus mud pumping and pot cover effect
Core Problem: Frost heave and thaw settlement in frozen high-speed railway subgrades are governed by coupled water and heat migration in the soil, and may be further intensified by traffic-induced vibration. The underlying hydro-mechanical processes in frozen, partially saturated subgrades remain poorly quantified.
Key Innovation: Identified a vibration-induced piston suction mechanism: cyclic vehicular loading acting on a frozen, low-permeability upper layer generated excess pore water pressure in the underlying unfrozen zone, establishing a sustained hydraulic gradient that pumped unfrozen water toward the freezing front.
49. Measuring ice lens initiation in soils by particle image velocimetry
Core Problem: Understanding the dynamic process of frost heave and ice lens formation in soils is limited by the lack of an adequate criterion.
Key Innovation: Using particle image velocimetry (PIV) to monitor frost heave and ice lens formation, proposing strain as a new criterion for assessing ice lens initiation across various soil types and freezing conditions.
50. Strain localization and time-dependent deformation in granodiorite characterized by distributed optical fiber sensing
Core Problem: Understanding the deformation process prior to macroscopic failure of brittle rocks, as well as the transient response during stress relaxation.
Key Innovation: Using distributed optical fiber sensing to measure strains across the sample surface, quantifying the degree of strain localization using the Gini coefficient.
51. A Back‐Trace Numerical Method for Calculating the Numerical Solution of the True Total Contributing Area for Real‐World Terrains
Core Problem: Existing flow direction algorithms perform poorly in Total Contributing Area (TCA) estimation due to irrational empirical strategies.
Key Innovation: A novel Back-Trace Numerical (BTN) method calculates the numerical solution of true TCA using Bicubic B-spline surface approximation, showing high accuracy on synthetic and real-world terrains.
52. Predicting Rainfall Infiltration Losses: A Rainfall Simulation Study of Land Cover, Slope and Soil Type
Core Problem: Rainfall-runoff models are calibrated using historical data to estimate loss parameters, which often deviate from physically realistic infiltration behavior.
Key Innovation: Developed prediction equations for rainfall infiltration losses based on physical attributes (grass cover, leaf litter, soil organic carbon, and bulk density) under controlled rainfall conditions.
53. An Integrated Machine Learning Approach for Real‐Time Prediction, Diagnostics and Optimization of Uranium‐Leaching Groundwater System
Core Problem: Traditional simulation and optimization of groundwater systems rely on process-based numerical models with low computational efficiency and unsatisfactory accuracy.
Key Innovation: A bidirectional simulation and optimization approach for adaptive prediction, diagnostics, and optimization of groundwater systems using recurrent and convolutional neural networks.
54. Winter Baseflow Calibration's Critical Role in Hydrological Modeling for the Pamir Region
Core Problem: Uncertainties in precipitation data greatly affect hydrological model accuracy in the Pamir Mountains, requiring multi-data calibration methods.
Key Innovation: Incorporating winter baseflow calibration alongside traditional calibration variables (runoff, snow cover, and glacier mass balance) reduces uncertainty ranges for snowmelt, glacier runoff, and baseflow estimates.
55. Unsupervised Characterization of Rain‐Induced Seismic Noise in Urban Fiber‐Optic Networks Using Deep Embedded Clustering
Core Problem: A physical model linking stormwater discharge processes to Distributed Acoustic Sensing (DAS)-recorded signals is lacking.
Key Innovation: A data-driven method, deep embedded clustering (DEC), to automatically detect and classify rain-induced noise from massive DAS data, predicting the presence of moderate to heavy rain and the duration of stormwater discharge.
56. InSAR Ground Deformation and Pumping Energy Consumption Reveal Urban Water Security
Core Problem: Water resource assessments are critical for ensuring water security, particularly in rapidly urbanizing regions with increasing water demand and limited water monitoring capabilities.
Key Innovation: Combining trends in pumping energy consumption and InSAR-derived ground deformation to assess water security in Cochabamba, Bolivia.
57. A Global Ensemble Forecast System (GEFS)-based synthetic event set of U.S. tornado outbreaks
Core Problem: Accurately assessing the risk of tornado outbreaks is difficult due to their rarity.
Key Innovation: Used weather model data to create hundreds of thousands of realistic but unseen tornado outbreak scenarios to estimate U.S. and local outbreak risk.
58. The first Earthquake Early Warning System for the high-speed railway in Italy: enhancing rapidness and operational efficiency during seismic events
Core Problem: Need for rapid and efficient response to seismic events to protect high-speed railway infrastructure.
Key Innovation: Developed an Earthquake Early Warning System that detects earthquakes in real time and sends alerts within seconds, slowing or stopping trains only where needed.
59. Meteorological Drought Trend Analysis and Forecasting Using a Hybrid SG-CEEMDAN-ARIMA-LSTM Model Based on SPI from Rain Gauge Data
Core Problem: Improving drought forecasting in uMkhanyakude, where water scarcity affects agriculture and livelihoods.
Key Innovation: A hybrid model combining Savitzky–Golay, decomposition methods, and neural networks showed high accuracy for early drought warning and water resource planning.
60. Hourly Precipitation Patterns and Extremization over Italy using convection-permitting reanalysis data
Core Problem: Understanding hourly precipitation patterns and extremes across Italy.
Key Innovation: Identified an increase in hourly extreme precipitation in Alpine areas during summer and southern coastal regions in autumn using the MERIDA HRES reanalysis dataset.
61. Ice thickness and subglacial topography of Swedish reference glaciers revealed by radio-echo sounding
Core Problem: Swedish glaciers are rapidly shrinking due to global warming, but their future evolution is uncertain.
Key Innovation: Radio-echo sounding was used to measure ice thickness and map bed topography of four Swedish glaciers, providing essential data for future projections.
62. Examining spin-up behaviour within WRF dynamical downscaling applications
Core Problem: Determining the spin-up time needed for a model's results to become independent of initial conditions.
Key Innovation: Comparing decadal simulations initialized at different times to determine convergence, suggesting at least one annual cycle is needed for spin-up in regional climate simulations.
63. Insights into uncertainties in future drought analysis using hydrological simulation model
Core Problem: Quantifying uncertainties in future runoff and drought projections due to parameter calibration.
Key Innovation: Using the Soil and Water Assessment Tool with multiple climate models and scenarios, the study highlights the significant impact of calibration choices on low-flow projections.
64. The impact of convection-permitting model rainfall on the dryland water balance
Core Problem: Most climate models cannot accurately represent dryland storms and their characteristics, which are key in controlling how water moves through the landscape.
Key Innovation: Using a simple hydrological model at four sites in the Horn of Africa, the study shows that using a model that can represent these storms results in higher soil moisture for plants and groundwater for humans.
65. Tracking space debris from sonic booms
Core Problem: Tracking the trajectory and break-up of space debris in Earth’s atmosphere is challenging.
Key Innovation: Using seismic data to reveal the trajectory and break-up of space debris as it enters Earth's atmosphere, providing a novel method for tracking these objects.
66. A multihazard assessment framework integrating single and coupled scenario analysis for droughts, floods, and landslides
Core Problem: Traditional hazard assessments often fail to account for the complex interactions between multiple natural hazards, leading to inaccurate risk evaluations.
Key Innovation: A multihazard assessment framework integrating single-hazard and coupled-hazard scenarios to identify coupling mechanisms and quantify their effects on composite hazard levels.
67. Experimental study on the influence of fissure spacing on the dynamic mechanical properties and energy evolution mechanisms of fissured rock
Core Problem: Understanding the influence of fissure spacing on the dynamic mechanical properties of rock masses is crucial for evaluating slope stability, but the mechanisms are not fully understood.
Key Innovation: Systematic investigation of the influence of fissure spacing on the dynamic mechanical behavior and failure mechanisms of rock masses using SHPB impact tests and high-speed camera recording.
68. Multi-Source Remote Sensing Data-Driven Susceptibility Mapping of Retrogressive Thaw Slumps in the Yangtze River Source Region
Core Problem: Lack of quantified assessment of the spatial occurrence probability of Retrogressive Thaw Slumps (RTSs) in the Yangtze River Source Region (YRSR).
Key Innovation: A susceptibility assessment framework using high-resolution optical remote sensing, time-series spectral features, and advanced ensemble learning algorithms to model the spatial distribution and risk levels of RTSs.
69. Remote Sensing, Vol. 18, Pages 333: An Optical–SAR Remote Sensing Image Automatic Registration Model Based on Multi-Constraint Optimization
Core Problem: Accurate registration of optical and SAR images is challenging due to radiometric discrepancies and geometric inconsistencies.
Key Innovation: An end-to-end Optical–SAR Registration Network (OSR-Net) based on multi-constraint joint optimization.
70. Remote Sensing, Vol. 18, Pages 325: Integrating AI for In-Depth Segmentation of Coastal Environments in Remote Sensing Imagery
Core Problem: Mapping coastal landforms is critical for sustainable management but requires detailed segmentation of surface types.
Key Innovation: Application of Transformer-based semantic segmentation models for pixel-level classification of coastal surface types.