Inventory, motion and acceleration of rock glaciers in Ile Alatau and Kungöy Ala-Too, northern Tien Shan, since the 1950s

Abstract

Abstract. Spatio-temporal patterns of rock glacier creep have rarely been studied outside the densely populated European Alps. This study investigates the spatial and temporal variability of rock glacier motion in the Ile Alatau and Kungöy Ala-Too mountain ranges, northern Tien Shan. Over the study region of more than 3000 km2, an inventory of slope movements is constructed using a large number of radar interferograms and high-resolution optical imagery. The inventory includes more than 900 landforms, of which around 550 are interpreted as rock glaciers. Out of the active rock glaciers, 45 are characterised by a rate of motion exceeding 1 m/a. From these fast rock glaciers we select six and study them in more detail (Gorodetzky, Morenny, Archaly, Ordzhonikidze, Karakoram and Kugalan Tash rock glaciers) using offset tracking between airphotos, and historical and modern very high resolution optical satellite data. Most of them show an overall increase of decadal surface velocities from the 1950s onwards with speeds being roughly two to three times higher in recent years compared to the 1950s and 1960s. This development indicates a possible significant increase in sediment and ice fluxes through rock glaciers and implies that – when compared to glacier shrinkage – periglacial sediment transport in the region seems to gain importance relative to glacial sediment transport. Those rock glacier fronts reaching the valley floors show a strongly compressive flow regime, and changes in speeds further upstream affect them only in a damped way. The only rock glacier investigated in detail that does not exhibit an overall increase in speed since the 1950s is Gorodetsky where glacier retreat and dead-ice degradation seem to have decoupled the rock glacier from its supply by glacial sediments and ice.

Permafrost in monitored unstable rock slopes in Norway – New insights from rock wall temperature monitoring, geophysical surveying and numerical modelling

“…The triggering of Veslemannen described in detail in Kristensen et al (2021) and the acceleration of the Adjet rock glacier (Eriksen et al, 2018) might be some of the first signs. For Gámanjunni-3, a rapid acceleration of the rock-glacier like landform forming the southern part of the rockslide, is possible, as described for various cases in the recent past in northern Norway (Eriksen et al, 2018), in the European Alps (Delaloye et al, 2008) and in Central Asia (Kääb et al, 2020). It could possibly lead to the triggering of secondary rock falls or debris flow, as described elsewhere (Lugon and Stoffel, 2010;Kummert et al, 2018).…”

Quantifying Overestimated Permafrost Extent Driven by Rock Glacier Inventory

“…Other inventories compiled by Schmid et al (2015), X. Wang et al (2017), Pandey (2019), and Kääb et al (2020) were not used because they did not provide details on RG state (intact or relict), and so were not sufficient to support the MOD RG estimation. In order to avoid problems with spatial auto-correlation and over-weighting of dense observations, point-scale permafrost presence and absence measurements were aggregated to a regular grid, co-registered with the unprojected SRTM30 grid (about 1 km pixel size).…”

Comment on tc-2021-77

“…The measurement of horizontal flow velocities of rock glaciers on remote sensing data, especially when using historical aerial images and their derivatives, is prone to errors. As described by Kääb et al (2020), the error budget is composed of the following components: 1) overall shifts between the orthorectified data 2) lateral shifts in the orthoimages due to errors in the DEM used for orthorectification 3) distortions in the aerial images or in the sensor model that propagate into the orthoimages 4) image matching uncertainties and errors. We minimized the shifts between the orthoimages by local co-registration of the orthoimages.…”

“…Errors of type 4) contain errors caused by the image correlation method itself. The measurement errors as consequence of image correlation vary with the image quality like resolution, shadow, contrast and noise of the image pairs (Kääb et al, 2020). We removed both directional and magnitudinal gross outliers manually by counterchecking the resulting displacement vectors with the corresponding orthoimage and hillshade pairs.…”