Rock glacier kinematics in the kaunertal, Ötztal alps, Austria

Abstract

The quantification of rock glacier kinematics on a regional basis has gained increasing importance in recent years. Here, we applied an image tracking approach on high-resolution aerial imagery to infer surface kinematics of 129 mapped rock glaciers in the Kaunertal, Austrian Alps. We find significant surface movement for 30 features with mean velocities falling between 0.11 and 0.29 m yr−1 and a maximum of 1.7 m yr−1. Local analysis and comparison to earlier studies reveals significant increases in rock glacier velocities in the study area. From the rock glacier inventory and high-resolution digital topography, we computed a series of morphometric parameters to analyze potential controls on rock glacier creep and to predict rock glacier activity using random forests and logistic regression models. The results point towards a stronger dependence of velocities on parameters describing general inclination, potentially acting as proxies for internal rock glacier properties, while activity states seem to be regulated mainly by rock glacier dimensions and topoclimate. Using a parameter subset, we successfully separated active from inactive rock glaciers with accuracies of up to 77.5%, indicating a promising approach to predict rock glacier activity solely relying on parameters that can be derived from regionally available data sets.