Detection and characterization of active landslides with multisource SAR data and remote sensing in western Guizhou, China

The western part of Guizhou is located in the second step East Asia. Although area stratigraphically continuous and surface dominated by hard limestone sandstone, catastrophic landslides often occur, seriously threatening residents' lives safety property. Accurate identification analysis their developmental patterns are vital to prevent reduce threat geological disasters. No active landslide survey data cover this region, so paper identifies combining deformation information, multitemporal optical remote sensing images, lithology, geomorphic features obtain information from multisource synthetic aperture radar data. This process increases accuracy reliability identifying unstable slopes areas with dense vegetation steep terrain. By processing 283 Sentinel-1 PALSAR-2 data, 578 landslides, 18 which high-risk large-scale (landslide groups), delineated for first time a range 4.64 × 104 km2 study area. mainly include natural reservoir mining-induced accounting 2.4%, 4.2 %, 93.4%, respectively. spatial distribution banded along cuesta at edge an outcrop coal strata. Landslides distributed elevations 1800–2000 m, elevation difference 50 ~ 100 m slope 35° 40°. characterized slopes, small scales, mass occurrences, no dominant direction, classifying them as induced mining disturbance. Furthermore, nuanced interpretation disaster elements, such cliff, tensile cracks, deep sizeable channels, isolated rock masses, collapse debris, processes change, reveals that mining-disturbed region have experienced four primary stages: unloading, disturbance, displacement acceleration, failure. great significance understanding genetic mechanism patterns, well risk assessment, region.