the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Nov 2025
Glacial decline next to stable permafrost in the Dry Andes? Vertical glacier surface changes and rock glacier kinematics based on Pléiades imagery (Rodeo basin, 2019–2025)
Abstract. The presence and volume of high-mountain cryospheric features are drastically affected by rising air temperatures – on global scale. In the Dry Andes, precipitation is extremely scarce, shifting the hydrological significance towards the solid water storages, glaciers and ground ice. While glaciers decrease in surface area and volume, periglacially stored waters, e.g., in rock glaciers, react more retarded to atmospheric forcing, potentially buffering future water availability. Despite rising air temperatures, recent studies suggest stable permafrost conditions in the Dry Andes based on borehole investigation and rock glacier kinematics for the last decade.
We investigate vertical surface changes of 19 glaciers, three debris-covered glaciers and 59 rock glaciers in the Rodeo basin (Dry Andes, Argentina) for the time period 2019–2025. Further, we calculate rock glacier velocities for 47 of the 59 rock glaciers for which we have data for all time periods. We follow photogrammetric principles using (tri)stereo, panchromatic Pléiades imagery to generate projected optical imagery and DEMs in Ames Stereo Pipeline that we co-register prior to DEM differencing for vertical surface change calculation. We conduct feature tracking on the panchromatic Pléiades imagery for the calculation of rock glacier velocities.
We detect glacier surface lowering of up to −8.99 m (cumulative, 2019–2025) and dominantly negative annual surface lowering for all glaciers investigated. We find vertical surface lowering on debris-covered glaciers to be well below the magnitude of glaciers but higher than for rock glaciers – the latter not exceeding a decimetre. We quantify rock glacier velocities of in average 0.28 to 0.82 m/yr (LoDs: ± 0.16, ± 0.61) and can categorize three rock glacier groups – large and fast, small and fast and small and slow. Across the 47 rock glaciers investigated, we do not find a regional trend of increasing velocities.
In conclusion, we observe a declining glacial domain to contrast with rock glacier velocities which elucidate stable permafrost conditions. We infer a delayed reaction of the periglacial domain to the rising temperatures that lead to the surface lowering of glaciers and highlight the need for ongoing, long(er)-time surface change monitoring in this crucial, dynamic point in time.
- Preprint
(2284 KB) - Metadata XML
-
Supplement
(1917 KB) - BibTeX
- EndNote
Status: open (until 05 Jan 2026)
- RC1: 'Comment on egusphere-2025-4630', Anonymous Referee #1, 01 Dec 2025 reply
-
RC2: 'Comment on egusphere-2025-4630', Dominik Amschwand, 15 Dec 2025
reply
First, I would like to apologize for the last-minute submission of my review. I enjoyed reading the paper and have no major issues, just a remark on the methodology, a suggestion for an additional analysis, and a few minor points.
The authors analyse vertical and horizontal surface velocities over the 6-year period 2019–2025 of 19 glaciers, 3 debris-covered glaciers, and 59 rock glaciers in the Rodeo catchment in the Dry Andes of Argentina using Pléiades imagery. The detailed kinematic analysis includes validation using in-situ DGNSS measurements at selected sites, which is in itself a logistical feat in such remote terrain. The main finding is that downwasting clean-ice and debris-covered glaciers, showing consistent and considerable surface lowering, contrast with kinematically stable rock glaciers, showing no trend (small vertical changes fluctuating about zero). The key contribution is a large kinematic data set in an otherwise data-scarce region at a crucial point of time, providing a baseline observation to assess the future permafrost evolution in the Rodeo catchment.
The important methodological choice in this work is to approach glaciers and rock glaciers, i.e., glacial and periglacial landforms, from the surface kinematics and the mass conservation equation, essentially arguing with the geodetic mass balance (as pioneered by Cusicanqui et al. (2021) for rock glaciers). In that sense, this study adds a valuable complementary approach to that of assessing changes in the ground thermal regime (as e.g., in Koenig et al. (2025)), where rock glaciers as climate-conditioned permafrost landforms are investigated with the lens of the surface energy balance/energy conservation equation. However, while down-wasting glaciers can be quite directly linked to climatic warming/drying (with little lag), kinematically stable rock glaciers are more indirect proxies of stable ground thermal conditions and the climatic forcing (L54–56, L548–556; Yu et al., 2025). The authors raise this point in L512ff: “The question here is the comparability between the meaning of vertical surface changes on glaciers and rock glaciers, which is why we focus on rock glacier velocities as indicator of (in)stability of permafrost conditions […]”. Depending on the ground temperature and the soil freezing characteristic curve (SFCC), ground warming might not lead to excess (!) ice melt and subsidence. Furthermore, interannual ice storage changes as estimated by Halla et al. (2021) for the Dos Lenguas rock glacier could in principle mask (in their short monitoring period) a slow long-term subsidence (such processes could be discussed more thoroughly). Taking horizontal creep rates into account is a smart move that makes the correspondence between stable ground thermal conditions and surface kinematics convincing enough in the scope of this analysis, albeit future in-situ investigations of the ground hydro-thermal regime would of course be helpful for a more conclusive assessment.
Minor comments and a suggestion for further analysis
Fig. 7. The large number panels are somewhat hard to synthesize for a human but would be doable for a machine. Consider, for example, a hierarchical clustering (of absolute or normalized median velocities), and report the (few) representative trend pattern(s). This would really be a useful (and not too costly) additional analysis to more quantitatively ground your key result (L541: “We do not detect a regional trend in increasing rock glacier velocities in the Rodeo basin between 2019-2025, Fig. 7”. (It would be interesting whether these clusters coincide with the three rock glacier groups “fast-large”, “fast-small”, and “small-slow/stagnant” as mentioned in L540f).
The vertical changes over the landforms are aggregated in terms of the median and not the mean (or any other measure). Why was the median chosen, and wouldn’t the mean value be more indicative of the whole-glacier geodetic mass balance? If the distribution of surface lowering is, say, right skewed, then the median is smaller than the mean and would underestimate the glacier-average changes. Are such considerations numerically relevant at all?
L25, “dominantly negative annual surface lowering for all glaciers investigated.” Consider writing “surface lowering” instead of “negative surface lowering”.
L190. LoD abbreviation not defined.
Fig. 7: The bars show a “fill level” and a colour. What do they mean? Colour coding: There are 4 colours (black, purple, blue, green) in Fig. 7, but only 3 in Fig. 5. What do the black lines refer to?
L309, L344, and others: Consider replacing “horizontal surface change” with “horizontal velocity” or “horizontal displacement (rates)” (the feature tracking gives labelled points whose position can be followed through time, not merely a change). This would make the terminology of “vertical surface change”, “rock glacier velocities” (always horizontally defined, correct?), and “glacier surface lowering” more consistent.
L533ff/L571. What do you mean by “volume dominated” vs. “creep-dominated” and “gravitational force to have a strong impact”? I do not fully understand this distinction, because all active rock glaciers move by gravity-driven creep. Do you refer to the material composition (ice content)? The motivation for this analysis could be better explained and better tied to the conclusions, currently it feels like an argumentative "dead end".
L540. Sentence unfinished (“Fig. 3C-D blue area on rock glacier front”).
L554: Possible additional references are works from A. Kellerer-Pirklbauer for the Austrian Alps, from PERMOS for the Swiss Alps, and from M. Marcer for the French Alps.
References
Cusicanqui, D., Rabatel, A., Vincent, C., Bodin, X., Thibert, E., & Francou, B. (2021). Interpretation of volume and flux changes of the Laurichard rock glacier between 1952 and 2019, French Alps. Journal of Geophysical Research: Earth Surface, 126, e2021JF006161. https://doi.org/10.1029/2021JF006161
Halla, C., Blöthe, J. H., Tapia Baldis, C., Trombotto Liaudat, D., Hilbich, C., Hauck, C., and Schrott, L.: Ice content and interannual water storage changes of an active rock glacier in the dry Andes of Argentina, The Cryosphere, 15, 1187–1213, https://doi.org/10.5194/tc-15-1187-2021, 2021.
Hock, R., Rasul, G., Adler, C., Cáceres, B., Gruber, S., Hirabayashi, Y., Jackson, M., Kääb, A., Kang, S., Kutuzov, S., Milner, Al., Molau, U., Morin, S., Orlove, B., and and H. Steltzer: High Mountain Areas, in: IPCC Special Report on the Ocean and Cryosphere in a Changing Climate, edited by: Pörtner, H.-O., Roberts, D. C., Masson-Delmotte, V., Zhai, P., M., T., Poloczanska, E., Mintenbeck, K., Alegría, A., Nicolai, M., Okem, A., Petzold, J., Rama, B., and Weyer, N. M., Cambridge University Press, 131–202, https://doi.org/10.1017/9781009157964.003, 2019.
Hu, Y., Arenson, L. U., Barboux, C., Bodin, X., Cicoira, A., Delaloye, R., et al. (2025). Rock glacier velocity: An essential climate variable quantity for permafrost. Reviews of Geophysics, 63, e2024RG000847. https://doi.org/10.1029/2024RG000847
Koenig, C. E. M., Hilbich, C., Hauck, C., Arenson, L. U., and Wainstein, P.: Thermal state of permafrost in the Central Andes (27–34° S), The Cryosphere, 19, 2653–2676, https://doi.org/10.5194/tc-19-2653-2025, 2025.
Citation: https://doi.org/10.5194/egusphere-2025-4630-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 174 | 99 | 27 | 300 | 36 | 20 | 19 |
- HTML: 174
- PDF: 99
- XML: 27
- Total: 300
- Supplement: 36
- BibTeX: 20
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comment
The paper presents a detailed investigation of glacial and, in particular, periglacial landforms in the Rodeo Basin of the Dry Andes. Using Pléiades stereo and tri-stereo imagery, the authors derive both vertical and horizontal surface displacements through DEM generation, DEM differencing, and feature-tracking techniques. They also provide an in-depth assessment of their methodology, including a rigorous quantification of landform velocities and the associated Levels of Detection (LoD).
The results indicate that the monitored rock glaciers show no consistent trend toward acceleration or deceleration. This suggests that permafrost-related landforms in the region currently exhibit a relatively stable deformation regime, highlighting their greater resilience to climate warming when compared with the faster melt rates observed for glaciers and debris-covered glaciers.
The paper is well structured, but some improvements and minor reorganizations could help make the text more fluent and clearer. Additionally, a broader discussion addressing some of the open issues (outlined in the following section) would strengthen the manuscript and contribute to a more comprehensive and robust overall presentation.
I suggest refining the writing in sections 4.5 and 5.3 to enhance fluency. Section 4.5 is presented largely as a sequence of results, and section 5.3 combines multiple themes, which affects the coherence of the argument.
Detailed comment
Abstract
The abstract is clear and presents the analyses conducted with a good level of detail. However, reading it, it sounds like a list of activities done, with maybe too many “we” in a row. E.g. “We investigate”, “we calculate”, “we follow”, “we conduct”, “we detect”, “we find” in only few lines. I would suggest revising it making the text more homogeneous and fluent.
Line 28: please insert the extended name of LoD acronym.
Line 130: Do the mapped rock glaciers already have an activity attribute (active/ inactive)? How is this attributed to them?
Line 170 and caption of Tab. 2: Stammler et a., YEAR (?)
Lines 177-178: It is not clear how the bounding box are created. First a buffer of 500m is extracted around glacier and rock glaciers polygons and then a bounding box is created around the so extracted features, right? Please try to clarify this better.
Lines 186-189: I find it quite difficult to understand the meaning of the cumulative median and how it is computed. Also, I don’t get if the vertical change is computed only as cumulative median over time or as vertical surface change normalized to full years. Could you please better formulate this sentence?
Lines 192-195: I suggest moving the section explaining how the LoD is derived right after the description of the image coregistration process (around line 186). This will make it immediately clear how the LoD is determined before you introduce the vertical surface-change quantification.
Lines 212: In the LoD estimation, the terrain outside the landform polygons is assumed stable (line 194). Yet, the text notes that the LoD for horizontal displacements accounts for potential true surface change (e.g., fluvial processes) occurring outside these polygons. Could you clarify this apparent inconsistency? If fluvial dynamics may be active, how is the ‘stable terrain’ assumption justified for LoD estimation?
Line 271: “an LoD”
Line 322- 327: The manuscript states that large and fast rock glaciers occur at higher elevations, on gentler slopes, and exhibit lower median vertical surface change. Could you elaborate on the physical mechanisms that might explain this pattern? If vertical deformation is limited, motion must be predominantly horizontal. What factors, beyond surface slope, could control this horizontal displacement? In addition, the relationships between elevation, slope, and the different rock-glacier classes are not immediately clear from the current figure. I suggest expanding this section and perhaps exploring a combined plot of elevation versus slope, with median velocity represented through a color gradient. Such a visualization could help clarify the spatial distribution and dynamics of the landforms.
Line 340: are these vertical or surface velocities? I would suggest always clarifying this in the text.
Lines 358-360: the sentence seems to miss a part, and it is not totally clear.
Lines 394-400: In these sentences, the differences between the DGNSS and Pléiades-based velocities are sometimes reported in meters (m), although velocities are expressed in m/yr. It is therefore unclear whether these values refer to differences in annualized velocities or to absolute positional offsets. Please clarify what these numbers represent and adjust the units or wording accordingly to avoid confusion.
Section 4.5: The section presents many numerical values (errors, medians, LoDs, differences, per year changes) for three sites and two time periods but rarely summarizes what these numbers mean.
Line 445: remove rock glaciers at the end of the sentence.
Line 494: by is repeated 2 times
Line 517: How were the UAV-derived surface changes obtained? From differences between point clouds or offset tracking on images?
Section 5.3: Overall, the section contains valuable comparisons and a solid contextualization of your vertical surface change results with previous studies. However, the text would benefit from clarification and improved structure to enhance readability. Several paragraphs are dense, and the narrative flow is sometimes difficult to follow. For example a reorganization of the text into smaller thematic paragraphs on snow cover impact, comparison with published mass balances and debris-coverd glaciers could make the text easier to read.
Figures:
Fig.2: What are the triangles, square and circles? It is not clear in the caption.
Fig. 3: Is the reported vertical surface change derived directly from the DoD between the 2019 and 2025 DEMs, or is it calculated as the cumulative sum of DoDs from consecutive DEMs?
Fig.5: As for Fig.2 it is not clear what squares, triangles and circles refer to. As also stated in the detailed comment (Line 322-327) I would suggest trying to plot elevation and slope in the same graph to see if the different types of rock glaciers cluster in specific areas.
Fig.7: In the vertical axis I would explicitly state “median horizontal velocity”
Fig.8: I suggest putting a title to this graph with also the position of the point considered.
Thanks