Predicting Landslides Susceptible Zones in the Lesser Himalayas by Ensemble of Per Pixel and Object-Based Models

Landslide susceptibility is a contemporary method for delineation of landslide hazard zones and holistically mitigating the future landslides risks planning decision-making. The significance this study that it would be first instance when ‘geon’ model will attempted to delineate map (LSM) complex lesser Himalayan topography as LSM technique. This adopted per-pixel-based ensemble approaches through modified frequency ratio (MFR) fuzzy analytical hierarchy process (FAHP) compared with ‘geons’ (object-based) aggregation produce an Kalsi-Chakrata road corridor. For models, 14 conditioning factors were carefully chosen; namely, slope, slope aspect, elevation, lithology, rainfall, seismicity, normalized differential vegetation index, stream power land use/land cover, soil, topographical wetness proximity drainage, road, fault. inventory data past derived from preceding satellite images, intensive field surveys, validation surveys. These divided into training test datasets following commonly accepted 70:30 ratio. GIS-based statistical techniques establish correlation between sites factors. To determine accuracy output, LSMs was validated methods receiver operating characteristics (ROC) relative density index (R-index). results indicate object-based geon produced higher (geon FAHP: 0.934; MFR: 0.910) over per-pixel (FAHP: 0.887; 0.841). noticeably showed constructs significant regional units mitigation strategies development. present may significantly benefit decision-makers planners in selecting appropriate risk procedures at local scale counter potential damages losses area.