EGU25-16592, updated on 15 Mar 2025
https://doi.org/10.5194/egusphere-egu25-16592
EGU General Assembly 2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Monday, 28 Apr, 08:30–10:15 (CEST), Display time Monday, 28 Apr, 08:30–12:30
 
Hall X2, X2.35
Monitoring rock rlacier kinematics with medium-resolution Landsat imagery: Insights from the Semi-arid Andes
Diego Cusicanqui1, Pascal Lacroix1, Xavier Bodin2, Benjamin Aubrey Robson3, Andreas Kääb4, and Shelley MacDonell5,6
Diego Cusicanqui et al.
  • 1ISTerre | CNES, Grenoble, France (diego.cusicanqui@univ-grenoble-alpes.fr)
  • 2Laboratoire EDYTEM, Univ. Savoie Mont-Blanc, Le Bourget du Lac, 73376, France
  • 3Department of Earth Science, University of Bergen, Bergen, Norway
  • 4Department of Geosciences, University of Oslo, Oslo, 0316, Norway
  • 5Centro de Estudios Avanzados en Zonas Áridas (CEAZA), La Serena, Chile
  • 6Waterways Centre, University of Canterbury and Lincoln University, Christchurch, New Zealand

Rock glacier velocity is increasingly recognised as a critical essential climatic variable (ECV) for monitoring permafrost dynamics. However, the representation of decadal regional spatio-temporal velocity patterns remains challenging due to the scarcity of high resolution (<5 m) remote sensing data. In contrast, mid-resolution (10-15 m) satellite imagery, such as Landsat, provides global coverage over several decades, but has not yet been systematically exploited for rock-glacier kinematics. This study presents a robust methodological framework that employs pairwise feature-tracking image correlation using medium-resolution optical Landsat 7/8 imagery. This method integrates surface displacement time series inversion and automatic persistent moving area (PMA) detection to monitor rock glacier activity in the semi-arid Andes of South America. Our approach enabled the detection and quantification of surface kinematics for 382 gravitational slope mass movements, of which 153 correspond to rock glaciers, over a 24-year period (1998-2022) and a study area of 2,250 km².

Remarkably, this is the first application of Landsat data to quantify rock glacier displacement and derive long-term velocity trends. The analysis reveals an average velocity of 0.37 ± 0.07 m m a-1 for all rock glaciers, with exceptional cases of large rock glaciers and debris-covered frozen landforms exhibiting surface velocities exceeding 2 m a-1. The results show good agreement with high-resolution optical imagery and recent in-situ measurements, although Landsat-derived velocities are systematically underestimated by about 20-30%. Furthermore, the relatively high uncertainties between consecutive image pairs pose a challenge for the interpretation of the annual velocity variations. Despite these limitations, our study identifies decadal velocity changes in 3% of PMAs, with three rock glaciers showing an 11% increase and six showing an 18% decrease in velocity over a decade. These results suggest a strong relationship between rock glacier velocity and physical controls such as size, slope, orientation and elevation. In particular, the results suggest that permafrost thaw significantly influences the spatial distribution of high-altitude landslides in the Andes, highlighting the role of cryospheric processes in landscape evolution.

This study demonstrates the feasibility of using medium-resolution optical imagery for global, long-term monitoring of rock glaciers, filling a critical gap in permafrost research. The application of such data offers unprecedented opportunities to improve our understanding of cryospheric dynamics and their implications for regional hydrology, geomorphology and climate change adaptation strategies.

How to cite: Cusicanqui, D., Lacroix, P., Bodin, X., Robson, B. A., Kääb, A., and MacDonell, S.: Monitoring rock rlacier kinematics with medium-resolution Landsat imagery: Insights from the Semi-arid Andes, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16592, https://doi.org/10.5194/egusphere-egu25-16592, 2025.