Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.18929451 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>This dataset provides an <strong>expert-interpreted inventory of moving areas</strong> mapped from <strong>InSAR deformation signals</strong> in high-mountain environments. Each feature corresponds to a polygon delineating a zone where the <strong>deformation signal is spatially coherent</strong> over the considered observation period (i.e., a “moving area” suitable for geomorphological interpretation). The inventory was compiled to support <strong>regional-scale mapping</strong>, <strong>process-based interpretation</strong>, and—critically—the <strong>training and benchmarking of machine-learning algorithms</strong> for automated detection/classification of moving terrain from InSAR products.<br><br>All mapped moving areas were <strong>geomorphologically interpreted</strong> (e.g., rock glaciers, deep-seated gravitational slope deformation, landslides, other moving surfaces where relevant). For each polygon, the dataset reports a <strong>velocity range / kinematic class</strong> derived from the InSAR signal and expressed following the <strong>Rock Glacier Inventories and Kinematics (RGIK) guidelines</strong> for handling kinematic information and moving areas on rock glaciers. Where classification is uncertain (e.g., low coherence, seasonal decorrelation, mixed signals), the dataset documents the uncertainty through conservative labeling and/or attribute flags (if included).<br><br>InSAR-derived motion is sensitive to <strong>geometry (LOS projection)</strong>, <strong>temporal decorrelation</strong>, and <strong>atmospheric artefacts</strong>, and may not capture rapid failures or motion orthogonal to the radar line-of-sight. Users should therefore treat the inventory as an <strong>interpreted product</strong> (not a complete census of all instabilities) and propagate uncertainties when using it for hazard assessment or model calibration.</p>