Thinning-induced glacier deceleration in the Zanskar Himalayas

Ghosh, Tirthankar; Ramsankaran, RAAJ; McCormack, Felicity S.; Mackintosh, Andrew N.

doi:10.5194/tc-20-3415-2026

Articles | Volume 20, issue 6

https://doi.org/10.5194/tc-20-3415-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/tc-20-3415-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 20, issue 6

Research article

|

16 Jun 2026

Research article |

| 16 Jun 2026

Thinning-induced glacier deceleration in the Zanskar Himalayas

Tirthankar Ghosh, RAAJ Ramsankaran, Felicity S. McCormack, and Andrew N. Mackintosh

Download

Final revised paper (published on 16 Jun 2026)
Supplement to the final revised paper
Preprint (discussion started on 14 Nov 2025)
Supplement to the preprint

Interactive discussion

Status: closed

CC1:
'Comment on egusphere-2025-5260', Daya Songh, 10 Dec 2025
General comments
The manuscript does not demonstrate sufficient novelty, methodological rigor, or scientific contribution to merit publication in its current form. Most findings reiterate well-established results from previous studies in the Himalaya, and the manuscript lacks a clearly articulated research gap, robust methodology, or quantitative analyses to substantiate its interpretations. Significant conceptual, methodological, and presentation issues further weaken the study.
Specific comments
Lack of Novelty and Insufficient Contribution
The manuscript largely repeats previously published results on glacier velocity, surface elevation change, and heterogeneous glacier response in the western Himalaya (e.g., Dehecq et al., 2019; Bhambri et al., 2023; Garg et al., 2025). The conclusions on glacier thinning, deceleration, debris-cover influence, and behaviour of lake-terminating glaciers are all well documented in existing literature and are not presented with new insights, methodological improvements, or conceptual advances.
The study area (Zanskar Basin) has already been extensively analysed (Rai et al., 2012; Pandey et al., 2012; Shukla & Qadir, 2016; Rana et al., 2023; Jasrotia et al., 2023; Mandal et al., 2024). The manuscript does not convincingly demonstrate how its results expand current understanding or address an unmet research need. The authors claim earlier works cover limited time periods, but Bhushan et al. (2018) computed regional mass budgets for 1999–2014, and Rana et al. (2023) produced velocity time series from 1999–2021. It is unclear what genuinely new temporal dataset or insight the present manuscript adds.
Issues in the Introduction
The Introduction incorrectly implies that earlier studies provide only short-term or two-epoch comparisons. However, prior works already assess long-term glacier changes. The manuscript must clarify what additional value its stated 1992–2023 velocity record brings, especially given the absence of field validation and without clear methodological innovation.
Study Area Description – Conceptual and Factual Errors
The authors state that they selected 12 glaciers: 11 valley glaciers and one cirque glacier (G12). However, G12 is misclassified, as it is a well-developed compound glacier system, not a cirque glacier.

The basis for selecting these 12 glaciers out of ~1700 in the basin is not explained. The authors must justify:

The total glacier count used for the basin, the criteria for choosing these specific glaciers, and how these 12 glaciers adequately represent the climatic and geometric diversity of the Zanskar region.

Figure 1 is confusing, with poorly defined basin boundaries and unclear representation of which glaciers actually lie within the Zanskar Basin.

If the goal is to derive generalizable interpretations, it is unclear why glaciers from different aspects, elevations, and climatic settings were not included.

Figures and Field Photographs

Figures 1b and 1c serve no scientific purpose as presented. They lack annotations, interpretation of field features, or justification for inclusion.

If fieldwork was performed, photographs of all studied glaciers or at least representative examples should be provided, not just two uncontextualized images.

Methodological Details Are Missing or Incomplete

The manuscript lacks critical methodological information necessary to evaluate scientific reliability:
The abstract and introduction mention “long-term velocity datasets (1992–2023)” but do not specify:

Satellite missions used,

Spatial/temporal resolutions,

Coregistration procedures,

Uncertainty estimation,

Filtering and post-processing of displacement vectors.

No validation is provided against ITS_LIVE, MEaSUREs, field data, or previous studies, raising serious concerns about accuracy.

Glacier boundary delineation and the strategy for sampling along flowlines are not described, and the location of the mentioned flowlines (e.g., Line 264) is not shown.

The claim of assessing “drivers of velocity change” is unsupported by any physical or statistical modelling.

The method for surface elevation change relies entirely on previously published datasets (Hugonnet et al., 2021) without explanation of why no DEM differencing was performed by the authors.

Overall, the methodological gaps are substantial and undermine confidence in the results.
Results – Conceptual Ambiguities and Incorrect Statements

The discussion comparing median velocities among glaciers is unclear and inconsistent. For example, G12 is characterised as a cirque glacier but then attributed low velocity to “geometry and ice thickness,” despite being a well-developed compound glacier.

Statements about lake-terminating glaciers exhibiting accelerations lack supporting analysis and do not engage with glacio-hydrological processes documented in the literature.

Weak Analysis and Overgeneralised Interpretation

The manuscript broadly attributes glacier deceleration to surface lowering and reduced driving stress without performing:

Stress-balance calculations,

Velocity–thickness scaling analyses,

Regression between thinning and velocity change.

Assertions regarding the influence of debris cover, glacier geometry, and lake termini remain anecdotal. No figures, quantitative assessments, or statistical metrics are provided.

Uncertainty estimates for velocity and elevation change are absent, making it impossible to judge the significance of the reported trends.

Conclusion
The manuscript suffers from fundamental issues in novelty, methodology, interpretation, and presentation. It reiterates previously established results without providing new scientific insight, lacks essential methodological transparency, and misinterprets or oversimplifies glacier dynamics.
Given the magnitude of conceptual, analytical, and structural shortcomings, the manuscript is not suitable for publication.
Citation: https://doi.org/10.5194/egusphere-2025-5260-CC1
- EC1: 'Editor statement: Unprofessional comment, lacking an affiliation', Etienne Berthier, 16 Dec 2025
  
  As editor handling this manuscript, I would like to express my concerns about the community comment posted by Daya Songh.
  First, both I and the co-editors-in-chief of The Cryosphere (TC) felt that the tone of the comment was unprofessional and goes beyond the level of negativity that would be deemed appropriate in peer review.
  Second, we would like to remind all colleagues that a pre-print considered for publication in TC “undergo(es) an interactive public discussion, during which the referees' comments (anonymous or attributed), additional community comments by other members of the scientific community (attributed), and the authors' replies are posted.” See https://publications.copernicus.org/services/public_peer_review.html
  Hence, we would please request that Daya Songh confirms their affiliation as we could not verify their identity, after which the authors can respond to the scientifically relevant points raised in this community comment.
  Etienne Berthier
  
  Citation: https://doi.org/10.5194/egusphere-2025-5260-EC1
- AC3: 'Reply on CC1', Tirthankar Ghosh, 18 Feb 2026
  
  Dear CC1,
  Thank you for your comments. You can find our response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5260-AC3
RC1:
'Comment on egusphere-2025-5260', Anonymous Referee #1, 27 Dec 2025

Summary
Ghosh et al. (2025) present a comprehensive analysis of long-term glacier surface velocity and elevation changes in the Zanskar Basin, Ladakh Himalaya. Utilizing satellite remote sensing data, primarily from Landsat imagery processed via COSI-Corr for velocity estimation and reanalyzed elevation data from Hugonnet et al. (2021), the study examines 12 selected glaciers to characterize trends in flow deceleration and thinning. The main findings indicate an overall velocity slowdown of -2.4 m yr⁻¹ decade⁻¹, with thinning accelerating from -0.22 m yr⁻¹ (2000–2005) to -0.57 m yr⁻¹ (2015–2020). The authors attribute these changes to climate-driven mass loss, which reduces gravitational driving stress and is modulated by glacier geometry, debris cover, and terminus type (e.g., lake-terminating glaciers exhibit localised accelerations). However, weaknesses were observed in inconsistent terminology, insufficient methodological details, and potential biases in ERA5 climatic data. The use of median rather than mean values for velocity and elevation changes may hinder comparisons with prior studies that use mean velocity and elevation changes. The discussion could be enhanced by incorporating regional contrasts, such as the accelerating surrounding Karakoram glaciers (Heid and Kääb, 2012), to better contextualise the slowdown of Zanskar’s glaciers. Please see my following suggestions. I hope this will help improve the text.
Suggestions:
L9: Several sentences include the term 'Glacier' twice or three times (L9, L16-19, L20, L23, L24, L40-41). For example, in this study, we have selected 12 glaciers as test glaciers to study the evolution of inter-annual glacier surface velocity and glacier surface elevation change (L110-111). This sentence contains the word 'glacier' four times.
L11: Remote Sensing provides a tool; what type of tool is it? You may mention the image correlation here.
L15: After point 1, a digit is sufficient, for example, -2.4 instead of -2.43 throughout the manuscript. Also, what does an uncertainty ± represent?
L15-16: Somewhere written year-1 and somewhere written yr-1. Please maintain consistency throughout the entire manuscript.
L16: Between 2000 and 2005, instead of 2000-2005. Please maintain consistency throughout the entire manuscript.
L28-29: Please use the word 'glacierized' instead of 'glaciated', as the latter word implies a former glacier area that has deglaciated now. Please check throughout the manuscript.
Please check this glossary- https://wgms.ch/downloads/Cogley_etal_2011.pdf
L35: You may replace 'impacted' with 'reduced'.
L38: Weertman (1957) is an outdated reference; you may replace it with more recent references.
L44: First mention the complete form of DGNSS and then write its acronym.
L45: Surface velocity can be estimated; a reference is needed.
L48: Two times mention 'particular' and 'particularly' in a single sentence.
L49-50: Satellite-based remote sensing methods. Please describe the specific techniques here.
L58: Please describe what other local conditions are.
L73: The three-year mass balance of Rulung Glacier and the four-year mass balance of Stok Glacier do not present a declining trend. To understand the trend in mass balance, we require a long series of in situ measurements. You may mention here the negative mass balance only.
L75-76: Correct the reference of Soheb et al. (2020) and Mehta et al. (2021).
L93: Omit Western Himalaya.
L97: Test……………………..reframe this sentence.
L110: Why were selected 12 glaciers? Please give the justification. What is the range of area and elevation of these glaciers? Please add 2-3 sentences on this.
L139: Green (B2 band), why was the Green band? Please explain here.
L158: There is some confusion. Did you co-register the Landsat images, as these images are already orthorectified? If not performed, please omit this.
L162: What is the "salt-and-pepper" effect?
L166: 4*4 pixels, but Landsat 5 does not contain a panchromatic band. Possibly you used 2*2.
L176-178: I feel five iterations are best, whereas two iterations may give some noise. Please carefully check and rewrite.
L204: It may also arise due to a horizontal shift in two images; you may mention this here.
L213: Please provide more details on pairwise uncertainty in the supplementary material.
L221-223: Please reframe this sentence, and replace glacier mass balance with elevation changes.
L229: Equation 5, how are these parameters calculated in this equation? What is the uncertainty in these estimates? Please describe in the text. I also could not find how this equation helps to interpret the result of glacier surface flow velocity in the manuscript.
L242: ERA5 Land reanalysis climate data: Many studies have reported biases in the ERA5 land reanalysis dataset in mountainous areas. Our own study found a 5- to 8-degree temperature difference between the automatic weather station and the ERA5 Land dataset in different seasons, without bias correction, in the Central Himalaya. Such climatic trend analysis without bias corrections is not appropriate and provides unrealistic trends. I would suggest conducting bias correction in ERA5 data using nearby IMD sites or the IMD grid dataset.
L247: Section 4.1, Glacier velocity trend and surface elevation change: Both sections must be presented separately. Also, section 4.3 "elevation-wise glacier velocity trend is actually section 4.2 (possibly typo mistake), which can be merged with section 4.1 (glacier velocity trends) following the removal of surface elevation change.
L257-258: In the result and discussion section for surface flow velocity and elevation changes, the median flow velocity term and median elevation change term are frequently used. However, most of the previous studies use mean or average surface flow velocity or mean elevation change values. It is not straightforward to compare the median surface flow velocity with the previously published mean surface flow velocity.
L260: Use 24.5 ±5.7 instead of 24.50±5.73 throughout the manuscript.
L263: T-statistic test, please describe this statistical test in detail in the methodology section for the interest of readers.
L281: Between 2001 and 2010, instead of between 2001-2010. Please carefully review the entire manuscript for any similar text.
L283: From 1992 to 2000 instead of from 1992-2000.
L296-300: This is related to methodology, not related to the result section. Please shift to the relevant section and write only the results here.
L326: omit theory.
L326: median glacier surface elevation change. Primarily, studies use mean glacier elevation changes. Please see my previous comment.
L327: What are bottom-heavy glaciers? First, define these terms.
L355; Figure 8: In sub-panel A, also show Drun Drung glacier in the subset, as shown in sub-figure b.
L380; Figure 9: In this figure, you mentioned median velocity and mean elevation, whereas in text L326, you mentioned median glacier surface elevation change. Please correct it throughout the manuscript.
L386; 4.4 Climatic trends: Please see my previous comment on the ERA5 Land dataset without Bias correction, such trends are not appropriate. Additionally, please describe how the calculated uncertainty of the ERA5 land dataset is ±0.01 °C in the methodology.
L396: The discussion section can also be improved. Please see the following papers, which can help enhance the discussion section, particularly regarding the increase in lake-terminating glacier flow velocity and its comparison with the Karakoram region.
Hyde, A., Carr, J.R., Dunning, S. and de Vries, M.V.W., 2025. Quantifying heterogeneous glacier dynamics in Lunana, Bhutan, using spatiotemporally high-resolution satellite imagery. Journal of Glaciology, pp.1-33.
Heid T, Kääb AJ. Repeat optical satellite images reveal widespread and long term decrease in land-terminating glacier speeds. The Cryosphere. 2012 Apr 10;6(2):467-78.
Wu, K., Liu, S., Xu, J., Zhu, Y., Liu, Q., Jiang, Z., and Wei, J., 2021. Spatiotemporal variability of surface velocities of monsoon temperate glaciers in the Kangri Karpo Mountains, southeastern Tibetan Plateau. Journal of Glaciology, 67(261), pp.186-191.

Citation: https://doi.org/10.5194/egusphere-2025-5260-RC1
- AC1: 'Reply on RC1', Tirthankar Ghosh, 15 Feb 2026
  
  Dear Anonymous Reviewer,
  Thank you for your comments. You can find our response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5260-AC1
RC2:
'Comment on egusphere-2025-5260', Maximillian Van Wyk de Vries, 10 Jan 2026

Please view the attached review file.

Citation: https://doi.org/10.5194/egusphere-2025-5260-RC2
- AC2: 'Reply on RC2', Tirthankar Ghosh, 15 Feb 2026
  
  Dear Dr Maximillian Van Wyk de Vries,
  Thank you for your comments. You can find our response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5260-AC2

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

ED: Publish subject to revisions (further review by editor and referees) (19 Feb 2026) by Etienne Berthier

AR by Tirthankar Ghosh on behalf of the Authors (13 Mar 2026) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (16 Mar 2026) by Etienne Berthier

RR by Maximillian Van Wyk de Vries (07 Apr 2026)

RR by Anonymous Referee #1 (18 Apr 2026)

Suggestions for revision or reasons for rejection

I sincerely apologize for the delay in completing my review. I thank the authors for their detailed reply and the revisions made. The revised manuscript is now scientifically strong and logically sound, and it fills an important gap in the understudied Zanskar region.
Most of my previous suggestions have been adequately incorporated. However, I would like to highlight one remaining issue related to the use of raw ERA5-Land data. The authors cite Mandal et al. (2024) to support the statistical performance of the raw dataset. While the coefficients of determination are indeed high (r² = 0.93–0.97 for temperature), Satyapragyan et al. (2026) demonstrate that strong linear correlations do not preclude large systematic biases. In the Higher Central Himalaya, raw ERA5/ERA5-Land showed seasonal temperature offsets of 5–8 °C and RMSEs of 7–8 °C. Notably, Mandal et al. (2024) also reported a wide RMSE range of 2.3–14.9 °C for temperature in their validation (Sect. 3.6), indicating that substantial absolute offsets exist despite the strong correlations.
To the best of my knowledge, trend analyses of raw versus bias-corrected ERA5 data can differ significantly (paper currently under review). Therefore, I strongly recommend that the authors clearly discuss this important limitation (Satyapragyan et al. 2026) in the Discussion section. This would greatly benefit readers and improve the transparency of the manuscript.
Minor suggestions:
L15: The velocity trend is given as “-2.4 m year⁻¹ decade⁻¹” in the abstract, but the text (L277) and Table 2 report -2.43. Please use consistent rounding (“-2.4” or “-2.43”) throughout.
L16: The abstract states approximately -0.21 m year⁻¹, whereas the text (L352) uses -0.22 m year⁻¹ for the 2000–2005 period. Please ensure consistency.
L276: The text reports a region-wide p-value of 0.027, whereas Table 2 shows p = 0.00 (i.e., p < 0.01) for “All Glaciers”. Please update the text to match the table or clarify that p = 0.027 refers to a different statistical test (e.g., pre- vs. post-2010).
L405: The reported winter precipitation decline of “-0.0003 mm yr⁻¹” is very small relative to the values shown in Figure 10b (~0.01–0.04 mm). Please consider rephrasing this as “no statistically significant change” for more clarity.
I hope this will help.
References:
Mandal, A., Vishwakarma, B. D., Angchuk, T., Azam, M. F., Garg, P. K., and Soheb, M.: Glacier mass balance and its climatic and nonclimatic drivers in the Ladakh region during 2000–2021 from remote sensing data, J. Glaciol., 70, e19, https://doi.org/10.1017/jog.2024.19, 2024
Satyapragyan, S., Yadav, J.S. and Bhambri R, (2026): Bias corrections of ERA5 and ERA5-land temperature using automatic weather station data in the Higher Central Himalaya: implications for hydro-meteorological and glaciological research. Journal of Hydrology: Regional Studies, 63, p.103079. https://doi.org/10.1016/j.ejrh.2025.103079

Hide

ED: Publish subject to minor revisions (review by editor) (28 Apr 2026) by Etienne Berthier

AR by Tirthankar Ghosh on behalf of the Authors (05 May 2026) Author's response Author's tracked changes

EF by Polina Shvedko (06 May 2026) Manuscript

ED: Publish subject to technical corrections (11 May 2026) by Etienne Berthier

AR by Tirthankar Ghosh on behalf of the Authors (16 May 2026) Manuscript

Short summary

Our long-term analysis (1992–2023) shows that glaciers in the Zanskar Himalayas are undergoing sustained flow deceleration, closely associated with progressive thinning and the resulting reduction in driving stress. Glacier-specific factors, including geometry, topography, debris cover, and terminus type, further modulate individual glacier response. Overall, these results provide new insight into the long-term dynamic response of glaciers in the region to ongoing mass loss.