Abstract
Permafrost is a condition that allows ice to form and persist in the ground because temperatures consistently remain below 0°C. In mountainous regions, this leads to the development of rock glaciers, which are masses of debris or fractured rocks mixed with ice. These formations move downslope due to gravity, influenced by the surrounding topography. Rock glaciers consist of angular, poorly sorted rocks embedded in glacial ice, which acts as a cement binding the boulders together. They are often characterized by steep fronts and distinct features such as lobate ridges and furrows, resulting from their viscous flow and various movement types, including sliding, rolling, and gravitational motion.
Monitoring rock glaciers is challenging due to their remote locations, harsh climatic conditions, and limited resources for operational systems. Traditional methods involve using GNSS instruments to measure the coordinates of specific objects, like boulders, to determine displacement and velocity. While these methods are highly accurate and capable of detecting 3D movements, they have limited spatial coverage and are time-consuming.
Remote sensing offers an alternative, particularly with Synthetic Aperture Radar (SAR), which enables the analysis of large areas and the detection of active zones, even under cloud cover. However, SAR has limitations, including challenges caused by surface roughness leading to signal scattering, difficulties in capturing vertical displacements, and limited resolution for detecting small debris and boulders.
To address these challenges, we propose integrating Unmanned Aerial Vehicles (UAVs) into a broader monitoring strategy that combines remote and ground-based sensing methods. UAVs can easily reach remote, rugged, and hazardous terrain, where it may be unsafe or impossible for humans to conduct surveys. UAVs equipped with various sensors, provide detailed data on terrain displacement, position, and characteristics. Outputs such as Digital Surface Models (DSMs), orthomosaics, thermal maps, and 3D point clouds are generated. Thermal imaging is particularly interesting because it reveals areas of higher activity or movement by detecting differences in surface temperature caused by internal friction or melting processes. While using the 3D cloud point data are then processed in modeling software, such as FLAC3D, to predict future scenarios and potential events, enhancing the understanding and management of rock glacier dynamics. We present this monitoring strategy in the rock glacier of Lazaun, in South Tyrol, Italy.