36 Advances in understanding landslides and other mass movements

In this session we want to highlight advances in understanding mass movements, including studies focused on description, mapping, monitoring, triggering mechanisms, mitigation, modeling, and evaluating susceptibility.

Wednesday, 11 March 2026
8:30 AM - 10:30 AM
RCC, 105

Landslide Susceptibility Modelling in a Structurally Complex Flysch Region: Weighted Overlay and Machine-Learning Analysis in the Latoritsa Basin, Ukrainian Carpathians

Paper number: 36-7
Program: Technical Sessions
Time: 10:05 AM – 10:20 AM

Authors

Abstract

Landslides represent one of the most significant natural hazards within the geodynamically active Ukrainian Carpathians, where complex geology, active tectonics, and intense climatic conditions contribute to slope instability. The Latoritsa River basin (Mukachevo district, Transcarpathian region) is particularly susceptible to landsliding due to structurally weakened Cretaceous-Paleogene flysch stratigraphy, highly dissected relief, and increasing frequency of extreme precipitation events. An early July 2025 rainfall episode triggered local flooding and reactivated numerous dormant slope instabilities, highlighting the growing relevance of landslide hazard assessment under changing climate conditions. This study aims to create a comprehensive landslide inventory through field mapping and to analyze the principal geological and geomorphological factors controlling slope failure processes in the basin.

A multi-source methodological framework was employed, integrating remote sensing data, field-based geological and geomorphological surveys, structural analysis, and GIS-based modelling. A detailed inventory of 206 mapped landslide sites was compiled, representing rotational and translational slides, debris flows, and rock falls. The central aspect of the research involved modelling landslide susceptibility through the combined analysis of dominant geomorphological and geological factors and the documented distribution of existing landslides. The analysis confirmed that landslide occurrence in the region exhibits strong spatial correlation with preexisting fault systems. This emphasizes the dominant role of tectonic processes and structural weaknesses in slope destabilization. Terrain parameters, including slope steepness, elevation, curvature, topographic wetness index, and moisture index, along with distance to rivers, distance to faults, lithology, precipitation, and distance to roads were incorporated as key conditioning factors for susceptibility modelling. These factors directly influence shear strength reduction, pore-water pressure dynamics, and stress redistribution within landslide-prone deposits.

Landslide susceptibility was first assessed using a traditional weighted overlay approach, where each reclassified raster was weighted according to its relative influence. To complement this method, an alternative dual machine-learning model leveraging a bagged trees parameter parser and a logistic regression was applied to evaluate non-linear relationships and improve predictive performance. The machine-learning approach provided enhanced discrimination of high-susceptibility zones, particularly in areas with complex terrain patterns. The results offer a robust scientific foundation for landslide hazard forecasting, risk assessment, and informed land-use planning. By integrating geological, geomorphological, and advanced geospatial modelling techniques, this study contributes essential insights toward improving early-warning systems and georisk mitigation strategies in the Carpathians.

 

Geological Society of America Abstracts with Programs. Vol. 58, No. 1

doi: 10.1130/abs/2026SE-411580

© Copyright 2026 The Geological Society of America (GSA), all rights reserved. Permission is hereby granted to the author(s) of this abstract to reproduce and distribute it freely, for noncommercial purposes. Permission is hereby granted to any individual scientist to download a single copy of this electronic file and reproduce up to 20 paper copies for noncommercial purposes advancing science and education, including classroom use, providing all reproductions include the complete content shown here, including the author information. All other forms of reproduction and/or transmittal are prohibited without written permission from GSA Copyright Permissions.