Abstract
Landslides represent one of the most widespread and impactful natural hazards, and their monitoring remains a key issue for risk assessment and mitigation. In recent years, multi-temporal Unmanned Aerial Vehicle (UAV) surveys have become an important data source for landslide monitoring, enabling detection of surface changes through repeated acquisitions. This study proposes a fully UAV-based workflow to quantify planar displacements of active, moderately rapid earthflows using multi-temporal Light Detection and Ranging (LiDAR) and optical data. The approach is tested on two landslides in the northern Apennines (Italy): the Sassi Neri and Antognano landslides, with different activity trends and geomorphological maturity. Three UAV campaigns were conducted for each landslide. Surface displacements were estimated using an automated offset tracking approach implemented through the SenseTrack algorithm, based on Digital Image Correlation applied to UAV-derived hillshade. The reliability of offset tracking results was validated through Homologous Point Tracking, based on manual tracking of surface objects and features on orthomosaics and hillshades. Displacement maps reveal a well-defined deformation pattern at Sassi Neri, whereas the Antognano landslide shows a more heterogeneous and discontinuous displacement distribution. Quantitative comparison between automated and manual measurements shows strong agreement for both case studies, with R² values of 0.92 for Sassi Neri and 0.93 for Antognano. These results demonstrate that multi-temporal UAV surveys combined with offset tracking provide accurate estimates of planar displacements from decimetric to multi-metric magnitudes. The proposed workflow is a rapid, flexible, and cost-effective solution for short- to medium-term landslide monitoring, particularly for emergency scenarios and areas where ground-based instrumentation is unviable.

This publication has 14 references indexed in Scilit: