the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Apr 2025
Increasing precipitation due to climate change could partially offset the impact of warming air temperatures on glacier loss in the monsoon-influenced Himalaya until 2100 CE
Abstract. Glacier volume in the Himalaya is projected to shrink by 53–70 % during this century due to climate change. However, the impact of changes in precipitation amount and distribution on future glacier change remains uncertain because mesoscale meteorology is not represented in current models that project glacier change. We explored the combined effects of past and future changes in air temperature and precipitation amount and distribution on the evolution of Khumbu Glacier in the Everest region of Nepal—a benchmark glacier in the monsoon-influenced Nepal Himalaya—using a climate-glacier modelling approach that forces an ice-dynamical glacier evolution model with a surface mass balance forcing that includes mesoscale meteorological variables derived from downscaling of Regional Climate Model outputs. Our simulations show that historical warming during the late Holocene has committed Khumbu Glacier to future volume loss of 10–23 % during this century. Under moderate future warming (RCP4.5) from the present day, Khumbu Glacier could lose 70 % volume by 2100 CE due to increasing air temperatures. However, the projected increase in precipitation in tandem with climate warming could offset half of this loss, such that the total decrease in glacier volume by 2100 CE compared to the present day is only 34 %. Extreme future warming (RCP8.5) will not be compensated by changes in precipitation but will instead result in substantial ablation above 6,000 m, causing the highest glacier on Earth to vanish between 2160–2260 CE.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(0 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1211', Emily Potter, 16 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1211/egusphere-2025-1211-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2025-1211-RC1 - AC1: 'Reply on RC1 and RC2', Ann Rowan, 16 Jun 2025
-
RC2: 'Comment on egusphere-2025-1211', Anonymous Referee #2, 20 May 2025
This is a review of the manuscript (no. egusphere-2025-1211) by Anya Schlich-Davies et al., entitled “Increasing precipitation due to climate change could partially offset the impact of warming air temperatures on glacier loss in the monsoon-influenced Himalaya until 2100 CE”, submitted for publication in The Cryosphere.
The authors investigate the mass balance and long-term evolution of Khumbu Glacier in the monsoon-influenced Himalaya from the Late Holocene through to 2300. Their approach combines multiple Regional Climate Models (CORDEX South Asia dataset) with two climate change scenarios (RCP4.5 and RCP8.5), a physics-based glacier surface energy and mass balance model (COSIPY), and a second-order ice flow model (iSOSIA). This integrated modelling framework is used to evaluate the effects of supraglacial debris, rising air temperatures, and changing precipitation patterns on future glacier mass loss. A notable strength of this study is its use of regional climate model data alongside the inclusion of key physical processes such as sublimation, avalanche-driven snow redistribution, and the evolution of the debris layer.
One of the central findings of the study is that under RCP4.5, projected mass loss of Khumbu Glacier due to global warming could be mitigated by up to 50% as a result of increasing precipitation trends projected by the considered climate models. The authors offer a valuable assessment of committed glacier mass loss from historical warming and outline potential future trajectories of this iconic glacier under different emission scenarios. These insights are not only relevant to the scientific community but are also critical for informing policy, particularly in the context of communicating the urgency and importance of emission reduction strategies to the public.
The modelling approach presented in the manuscript is ambitious and holds significant promise for improving the robustness and reliability of future projections for (debris-covered) glaciers in high mountain environments. However, in my view, the manuscript requires substantial and careful revision. In particular, key aspects of the methodology lack sufficient detail and clarity, which limits the transparency and reproducibility of the results. I outline these concerns in more detail below.
Aim of the Study and Rationale for the Modelling Approach
The manuscript would benefit from the clear formulation of a central research question - e.g., How will projected changes in precipitation and debris thickness affect the mass balance and long-term evolution of Khumbu Glacier? - or the articulation of a testable hypothesis. While the overall aim becomes apparent throughout the manuscript, it is not stated explicitly, either in the abstract or the introduction. Including a concise statement of objectives would help guide the reader and clarify the study’s focus from the outset.
In the introduction, I would expect a dedicated paragraph summarising previous modelling efforts for (debris-covered) glaciers in the region, along with a discussion of their limitations. This would provide context and a stronger motivation for the implementation of the enhanced modelling framework proposed in this study. The choice to extend the modelling period beyond 2100 is appreciated and commendable, as it allows for exploration of long-term glacier response to historical and projected climate forcing. However, the rationale behind selecting the specific modelling period (Late Holocene to 2300 CE) as well as the choice of models (COSIPY and iSOSIA) should be better justified and discussed.
The authors assert that their modelling approach allows for a more robust representation of mesoscale meteorological phenomena compared to previous studies. However, they neither specify which phenomena are meant nor explain why these would be better captured by the selected Regional Climate Models (RCMs), which operate at a horizontal resolution of 50 km - still insufficient to resolve the complex orographic and microclimatic variability of the Everest region. Recognition is due for the integration of important physical processes such as avalanche-driven snow redistribution, sublimation, and debris layer evolution. Nevertheless, it is somewhat disappointing that the individual effects of these processes on the surface energy and mass balance, as well as on glacier evolution, are not discussed in more detail. Instead, they are only briefly mentioned in a single sentence (Lines 478-479): “Our results show that avalanching and sublimation are important controls on recent and future glacier evolution for Khumbu Glacier.” This deserves a more in-depth treatment.
Terminology
I recommend that the authors carefully review and revise the terminology used throughout the manuscript to ensure consistency and scientific accuracy. A few examples of inconsistent or suboptimal phrasing include:
Line 6: “warming air temperatures” => replace with increasing or rising air temperatures.
Lines 34 & 83: “highest glacier on Earth” => this probably depends on the exact definition (mean/max/min elevation), stating just the elevation range as you did is sufficient. Instead of the superlative, I would rather emphasise the wealth of observational data available for Khumbu Glacier as the main rationale for its selection.
Line 69: “climate warming” => global warming or climate change
Lines 30 & 44: “moderate warming scenario RCP4.5” => given that RCP4.5 projects a global mean warming of approximately 2.5-3.0 °C, it is more appropriate to describe it as an “intermediate” scenario, in line with IPCC terminology.
Line 45: “extreme warming scenario RCP8.5” => see comment before and consider rather “high” or “pessimistic” emission scenario.
Data and Methods: Observations and Remote Sensing Datasets
A concise overview of the glaciological and meteorological observations, as well as remote sensing datasets used in the study, would be helpful to better understand the evaluation of the climate data and modelling results right from the beginning. I suggest including a section within the methodology, potentially supported by a summary Table (the location of key measurements could be indicated in Fig. 1c).
Data and Methods: Bias Correction and Statistical Downscaling of RCM Outputs
While the general workflow for bias correction and statistical downscaling is understandable, several important aspects remain unclear and merit further elaboration:
- Lines 122-124: “Three of the six available CORDEX South Asia RCMs (NOAA, CCCma, IPSL) were selected as discrete scenarios that span the range of possible future precipitation conditions (Table 1); either wet, moderate, or dry climate in 2080–2100 CE”. What were the selection criteria for these three models? Please clarify how “wet,” “moderate”, and “dry” future climates are defined. Ideally, provide quantitative precipitation trends for each case to substantiate this classification.
- Figure A3: If I interpret this figure correctly, all three selected RCMs significantly overestimate mean annual precipitation (by up to 500%; i.e., 3000 mm vs. observed 600 mm/year) and the fraction of non-monsoonal precipitation. If so, this casts doubt on the reliability of the RCM projections. Does quantile mapping preserve the original projected precipitation trends? Please provide a brief discussion of the resulting uncertainties and potential implications for glacier modelling.
- Lines 408-412: The sentence describing the use of AWS data and time-slice selection is unclear. How exactly did you use 14 years of AWS data to downscale and bias-correct five years of RCM outputs? This requires clearer phrasing and more detailed explanation.
- Spatial input fields: How were spatial fields for air temperature and precipitation generated for the SEB/SMB simulations? Did you simply apply gradients (if yes, how were they calculated) in combination with the resampled 100-m DEM?
- RCM variables for COSIPY: Please include a Table listing all RCM-derived variables (with units) used in the SEB/SMB simulations, indicating which variables were bias-corrected or downscaled (beyond temperature and precipitation, if applicable).
- Table 1: It is unclear why the projected temperature increase from 2100 to 2200 is considerably lower than in the previous and following centuries for both scenarios. This warrants clarification. Also, please correct what appears to be a typo in the last column header. It should refer to changes relative to 2200, not 2300.
Data and Methods: Glacier Modelling (Sections 2.3, 2.4 and 3)
Please note that Section 3 exists twice (Line 243 and 273). Although I have read the glacier modelling sections multiple times, I still found the workflow difficult to follow without referring back to Rowan et al. (2015). For clarity and completeness, I recommend addressing the following points:
- Glacier evolution model summary: While the glacier dynamics model has been previously described in detail by Rowan et al. (2015), the current manuscript would benefit from a concise summary of its key principles and the calibration approach. For example, was the model calibrated against observed surface velocities, ice thickness distributions, or terminus positions? How was model performance evaluated?
- Methodological advancements over prior work: The authors should clearly identify the specific methodological advancements of this study relative to their previous publications (e.g. Rowan et al., 2015, 2021). Is the integration of COSIPY the main novel component? Does the use of RCM-forced surface energy balance modelling represent a substantial improvement in boundary conditions for the dynamic model?
- Clarification of “late Holocene” reference: The “late Holocene” is mentioned as starting point for the simulations at different sections in the manuscript. Please provide a consistent and specific date (e.g. “1.3 ± 0.1 ka”) early on in the text. Figure 1c should also show the location of the ice-marginal moraines used to constrain the glacier extent and thickness during the spin-up period. Additionally, it remains unclear which climate data were used for the 5000-year equilibrium simulation (Line 244) and how the transition to the Little Ice Age (LIA) was handled. How is the LIA defined in this context for Khumbu Glacier?
- Choice and calibration of COSIPY: Please include a brief explanation why you chose COSIPY for your SMB simulations? The great advantage of COSIPY is that it considers physical processes such as sublimation and refreezing that are relevant in monsoon-influenced settings, but the drawback is that it is computatiolly costly and highly sensitive to climate input data and parameter calibration (see e.g. Temme et al., 2023). I could not to find any information on model calibration or any kind of sensitivity test. Did you calibrate the model against glaciological or geodetic mass balance observations? Which values (default?) did you use for the various parameters (e.g. roughness length, albedo) in COSIPY? Please include a table with the model parameters and their values (or ranges if calibrated). Also add uncertainties to your SMB simulations (Fig. 3 and Fig. 7B). Have you tested to run COSIPY with 3-hourly or 6-hourly time steps to reduce computational costs? Moreover, it would be valuable to visualise and discuss key physical fluxes simulated by COSIPY - e.g., trends in energy components (net radiation, turbulent fluxes), the role of sublimation, meltwater retention/refreezing - to support and contextualise your conclusions.
- Coupling of COSIPY and iSOSIA: Based on Figure 2A and the manuscript text, it appears that COSIPY and iSOSIA were run in a one-way rather than fully coupled manner. Please clarify. How often (e.g. every season, year, decade) did you update the SMB in the ice dynamics model to compute avalanching, ice flow etc.? Wich SMB did you use between the two time slices (2015-2020 and 2095-2100) to inform the ice dynamics model? Was glacier geometry (e.g. surface elevation) updated in COSIPY using information from iSOSIA (if yes, at which time steps?) to account for SMB-elevation feedbacks?
- Precipitation and glacier volume evolution after 2100: The glacier volume trajectories in Fig. 7B show abrupt changes, particularly under RCP8.5, that appear to correspond to significant changes in precipitation (trends). This raises questions about the plausibility of the long-term forcing. Please include a figure (e.g. in the appendix) showing the evolution of multi-annual or decadal mean temperature and precipitation for the full simulation period (2000–2300) for each RCM/RCP combination. Discuss how reliable these projections are.
Presentation and Discussion of Findings (Sections 3, 4 and 5)
- I found it difficult to interpret the 30 different 2.5 D visualisation of glacier mass balance, ice thickness and debris thickness in Figs. 4, 5 and 7. I would suggest to show the results for the best-performing RCM (in Fig. 4; move the remaining tiles to the appendix), increase the the size of individual tiles and the legends, use different colour schemes for different variables and add a reference outline (e.g. “present day”) in all simulation panels for easier comparison. In addition, a summarising table comparing the performance of different model setups and their agreement with observational data would be very helpful. If feasible, an animation (e.g. a GIF showing the glacier’s temporal evolution in terms of mass balance, geometry, and debris cover) could be a valuable supplement and make the results more accessible to both scientific and non-scientific audiences.
- One of the central findings is the potential mitigation effect of an projected increase on the long-term glacier mass balance. If this remains a key focus of the manuscript, I would expect a deeper and more critical discussion of (i) how robust these projected precipitation trends are across the different RCMs and (ii) how these findings compare with other studies and what implications they have for our broader understanding of the future evolution of glaciers in this region.
- The importance of sublimation and avalanching for the modelling of the future evolution of Khumbu Glacier is emphasised (Line 478-479). However, the current discussion is too brief. I would encourage the authors to support their claims with quantitative evidence or a figure.
- The comparison with the global glacier modelling results of Rounce et al. (2023) in Fig. 8 is a valuable addition. However, please use a consistent reference date (e.g. 2010) for both your model simulations and the global study. I would also suggest to state mass losses instead of volume losses for easier comparison with other studies (e.g. Kraaijenbrink et al., 2017; Rounce et al., 2023). The 39% difference in projected mass loss for Khumbu Glacier under RCP4.5 between this study and Rounce et al. (2023) is substantial and merits a more thorough discussion. What are the key drivers of this discrepancy - differences in SMB model physics, debris representation, regional precipitation trends, or calibration strategies?
Additional references that might be of interest for the introduction and discussion
Arndt & Schneider (2023): https://doi.org/10.1017/jog.2023.46 => Modelling glacier mass balance and investigating sensitivity to atmospheric forcing in High Mountain Asia
Collier et al. (2013): https://doi.org/10.5194/tc-7-779-2013 => Interactive modelling of glacier-atmosphere interactions
Collier et al. (2015): https://doi.org/10.5194/tc-9-1617-2015 => Modelling the impact of debris cover on ablation and glacier-atmosphere feedbacks
Compagno et al. (2022): https://doi.org/10.5194/tc-16-1697-2022 => Modelling supraglacial debris-cover evolution
Temme et al. (2023): https://doi.org/10.5194/tc-17-2343-2023 => Strategies for SMB model calibration (including COSIPY)
Zekollari et al. (2017): https://doi.org/10.5194/tc-11-805-2017 => Complete different setting, but insightful simulations and discussion on the potential mitigation impact of increasing precipitation under different warming scenarios (see e.g. Fig. 11)
Citation: https://doi.org/10.5194/egusphere-2025-1211-RC2 - AC1: 'Reply on RC1 and RC2', Ann Rowan, 16 Jun 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1211', Emily Potter, 16 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1211/egusphere-2025-1211-RC1-supplement.pdf
- AC1: 'Reply on RC1 and RC2', Ann Rowan, 16 Jun 2025
-
RC2: 'Comment on egusphere-2025-1211', Anonymous Referee #2, 20 May 2025
This is a review of the manuscript (no. egusphere-2025-1211) by Anya Schlich-Davies et al., entitled “Increasing precipitation due to climate change could partially offset the impact of warming air temperatures on glacier loss in the monsoon-influenced Himalaya until 2100 CE”, submitted for publication in The Cryosphere.
The authors investigate the mass balance and long-term evolution of Khumbu Glacier in the monsoon-influenced Himalaya from the Late Holocene through to 2300. Their approach combines multiple Regional Climate Models (CORDEX South Asia dataset) with two climate change scenarios (RCP4.5 and RCP8.5), a physics-based glacier surface energy and mass balance model (COSIPY), and a second-order ice flow model (iSOSIA). This integrated modelling framework is used to evaluate the effects of supraglacial debris, rising air temperatures, and changing precipitation patterns on future glacier mass loss. A notable strength of this study is its use of regional climate model data alongside the inclusion of key physical processes such as sublimation, avalanche-driven snow redistribution, and the evolution of the debris layer.
One of the central findings of the study is that under RCP4.5, projected mass loss of Khumbu Glacier due to global warming could be mitigated by up to 50% as a result of increasing precipitation trends projected by the considered climate models. The authors offer a valuable assessment of committed glacier mass loss from historical warming and outline potential future trajectories of this iconic glacier under different emission scenarios. These insights are not only relevant to the scientific community but are also critical for informing policy, particularly in the context of communicating the urgency and importance of emission reduction strategies to the public.
The modelling approach presented in the manuscript is ambitious and holds significant promise for improving the robustness and reliability of future projections for (debris-covered) glaciers in high mountain environments. However, in my view, the manuscript requires substantial and careful revision. In particular, key aspects of the methodology lack sufficient detail and clarity, which limits the transparency and reproducibility of the results. I outline these concerns in more detail below.
Aim of the Study and Rationale for the Modelling Approach
The manuscript would benefit from the clear formulation of a central research question - e.g., How will projected changes in precipitation and debris thickness affect the mass balance and long-term evolution of Khumbu Glacier? - or the articulation of a testable hypothesis. While the overall aim becomes apparent throughout the manuscript, it is not stated explicitly, either in the abstract or the introduction. Including a concise statement of objectives would help guide the reader and clarify the study’s focus from the outset.
In the introduction, I would expect a dedicated paragraph summarising previous modelling efforts for (debris-covered) glaciers in the region, along with a discussion of their limitations. This would provide context and a stronger motivation for the implementation of the enhanced modelling framework proposed in this study. The choice to extend the modelling period beyond 2100 is appreciated and commendable, as it allows for exploration of long-term glacier response to historical and projected climate forcing. However, the rationale behind selecting the specific modelling period (Late Holocene to 2300 CE) as well as the choice of models (COSIPY and iSOSIA) should be better justified and discussed.
The authors assert that their modelling approach allows for a more robust representation of mesoscale meteorological phenomena compared to previous studies. However, they neither specify which phenomena are meant nor explain why these would be better captured by the selected Regional Climate Models (RCMs), which operate at a horizontal resolution of 50 km - still insufficient to resolve the complex orographic and microclimatic variability of the Everest region. Recognition is due for the integration of important physical processes such as avalanche-driven snow redistribution, sublimation, and debris layer evolution. Nevertheless, it is somewhat disappointing that the individual effects of these processes on the surface energy and mass balance, as well as on glacier evolution, are not discussed in more detail. Instead, they are only briefly mentioned in a single sentence (Lines 478-479): “Our results show that avalanching and sublimation are important controls on recent and future glacier evolution for Khumbu Glacier.” This deserves a more in-depth treatment.
Terminology
I recommend that the authors carefully review and revise the terminology used throughout the manuscript to ensure consistency and scientific accuracy. A few examples of inconsistent or suboptimal phrasing include:
Line 6: “warming air temperatures” => replace with increasing or rising air temperatures.
Lines 34 & 83: “highest glacier on Earth” => this probably depends on the exact definition (mean/max/min elevation), stating just the elevation range as you did is sufficient. Instead of the superlative, I would rather emphasise the wealth of observational data available for Khumbu Glacier as the main rationale for its selection.
Line 69: “climate warming” => global warming or climate change
Lines 30 & 44: “moderate warming scenario RCP4.5” => given that RCP4.5 projects a global mean warming of approximately 2.5-3.0 °C, it is more appropriate to describe it as an “intermediate” scenario, in line with IPCC terminology.
Line 45: “extreme warming scenario RCP8.5” => see comment before and consider rather “high” or “pessimistic” emission scenario.
Data and Methods: Observations and Remote Sensing Datasets
A concise overview of the glaciological and meteorological observations, as well as remote sensing datasets used in the study, would be helpful to better understand the evaluation of the climate data and modelling results right from the beginning. I suggest including a section within the methodology, potentially supported by a summary Table (the location of key measurements could be indicated in Fig. 1c).
Data and Methods: Bias Correction and Statistical Downscaling of RCM Outputs
While the general workflow for bias correction and statistical downscaling is understandable, several important aspects remain unclear and merit further elaboration:
- Lines 122-124: “Three of the six available CORDEX South Asia RCMs (NOAA, CCCma, IPSL) were selected as discrete scenarios that span the range of possible future precipitation conditions (Table 1); either wet, moderate, or dry climate in 2080–2100 CE”. What were the selection criteria for these three models? Please clarify how “wet,” “moderate”, and “dry” future climates are defined. Ideally, provide quantitative precipitation trends for each case to substantiate this classification.
- Figure A3: If I interpret this figure correctly, all three selected RCMs significantly overestimate mean annual precipitation (by up to 500%; i.e., 3000 mm vs. observed 600 mm/year) and the fraction of non-monsoonal precipitation. If so, this casts doubt on the reliability of the RCM projections. Does quantile mapping preserve the original projected precipitation trends? Please provide a brief discussion of the resulting uncertainties and potential implications for glacier modelling.
- Lines 408-412: The sentence describing the use of AWS data and time-slice selection is unclear. How exactly did you use 14 years of AWS data to downscale and bias-correct five years of RCM outputs? This requires clearer phrasing and more detailed explanation.
- Spatial input fields: How were spatial fields for air temperature and precipitation generated for the SEB/SMB simulations? Did you simply apply gradients (if yes, how were they calculated) in combination with the resampled 100-m DEM?
- RCM variables for COSIPY: Please include a Table listing all RCM-derived variables (with units) used in the SEB/SMB simulations, indicating which variables were bias-corrected or downscaled (beyond temperature and precipitation, if applicable).
- Table 1: It is unclear why the projected temperature increase from 2100 to 2200 is considerably lower than in the previous and following centuries for both scenarios. This warrants clarification. Also, please correct what appears to be a typo in the last column header. It should refer to changes relative to 2200, not 2300.
Data and Methods: Glacier Modelling (Sections 2.3, 2.4 and 3)
Please note that Section 3 exists twice (Line 243 and 273). Although I have read the glacier modelling sections multiple times, I still found the workflow difficult to follow without referring back to Rowan et al. (2015). For clarity and completeness, I recommend addressing the following points:
- Glacier evolution model summary: While the glacier dynamics model has been previously described in detail by Rowan et al. (2015), the current manuscript would benefit from a concise summary of its key principles and the calibration approach. For example, was the model calibrated against observed surface velocities, ice thickness distributions, or terminus positions? How was model performance evaluated?
- Methodological advancements over prior work: The authors should clearly identify the specific methodological advancements of this study relative to their previous publications (e.g. Rowan et al., 2015, 2021). Is the integration of COSIPY the main novel component? Does the use of RCM-forced surface energy balance modelling represent a substantial improvement in boundary conditions for the dynamic model?
- Clarification of “late Holocene” reference: The “late Holocene” is mentioned as starting point for the simulations at different sections in the manuscript. Please provide a consistent and specific date (e.g. “1.3 ± 0.1 ka”) early on in the text. Figure 1c should also show the location of the ice-marginal moraines used to constrain the glacier extent and thickness during the spin-up period. Additionally, it remains unclear which climate data were used for the 5000-year equilibrium simulation (Line 244) and how the transition to the Little Ice Age (LIA) was handled. How is the LIA defined in this context for Khumbu Glacier?
- Choice and calibration of COSIPY: Please include a brief explanation why you chose COSIPY for your SMB simulations? The great advantage of COSIPY is that it considers physical processes such as sublimation and refreezing that are relevant in monsoon-influenced settings, but the drawback is that it is computatiolly costly and highly sensitive to climate input data and parameter calibration (see e.g. Temme et al., 2023). I could not to find any information on model calibration or any kind of sensitivity test. Did you calibrate the model against glaciological or geodetic mass balance observations? Which values (default?) did you use for the various parameters (e.g. roughness length, albedo) in COSIPY? Please include a table with the model parameters and their values (or ranges if calibrated). Also add uncertainties to your SMB simulations (Fig. 3 and Fig. 7B). Have you tested to run COSIPY with 3-hourly or 6-hourly time steps to reduce computational costs? Moreover, it would be valuable to visualise and discuss key physical fluxes simulated by COSIPY - e.g., trends in energy components (net radiation, turbulent fluxes), the role of sublimation, meltwater retention/refreezing - to support and contextualise your conclusions.
- Coupling of COSIPY and iSOSIA: Based on Figure 2A and the manuscript text, it appears that COSIPY and iSOSIA were run in a one-way rather than fully coupled manner. Please clarify. How often (e.g. every season, year, decade) did you update the SMB in the ice dynamics model to compute avalanching, ice flow etc.? Wich SMB did you use between the two time slices (2015-2020 and 2095-2100) to inform the ice dynamics model? Was glacier geometry (e.g. surface elevation) updated in COSIPY using information from iSOSIA (if yes, at which time steps?) to account for SMB-elevation feedbacks?
- Precipitation and glacier volume evolution after 2100: The glacier volume trajectories in Fig. 7B show abrupt changes, particularly under RCP8.5, that appear to correspond to significant changes in precipitation (trends). This raises questions about the plausibility of the long-term forcing. Please include a figure (e.g. in the appendix) showing the evolution of multi-annual or decadal mean temperature and precipitation for the full simulation period (2000–2300) for each RCM/RCP combination. Discuss how reliable these projections are.
Presentation and Discussion of Findings (Sections 3, 4 and 5)
- I found it difficult to interpret the 30 different 2.5 D visualisation of glacier mass balance, ice thickness and debris thickness in Figs. 4, 5 and 7. I would suggest to show the results for the best-performing RCM (in Fig. 4; move the remaining tiles to the appendix), increase the the size of individual tiles and the legends, use different colour schemes for different variables and add a reference outline (e.g. “present day”) in all simulation panels for easier comparison. In addition, a summarising table comparing the performance of different model setups and their agreement with observational data would be very helpful. If feasible, an animation (e.g. a GIF showing the glacier’s temporal evolution in terms of mass balance, geometry, and debris cover) could be a valuable supplement and make the results more accessible to both scientific and non-scientific audiences.
- One of the central findings is the potential mitigation effect of an projected increase on the long-term glacier mass balance. If this remains a key focus of the manuscript, I would expect a deeper and more critical discussion of (i) how robust these projected precipitation trends are across the different RCMs and (ii) how these findings compare with other studies and what implications they have for our broader understanding of the future evolution of glaciers in this region.
- The importance of sublimation and avalanching for the modelling of the future evolution of Khumbu Glacier is emphasised (Line 478-479). However, the current discussion is too brief. I would encourage the authors to support their claims with quantitative evidence or a figure.
- The comparison with the global glacier modelling results of Rounce et al. (2023) in Fig. 8 is a valuable addition. However, please use a consistent reference date (e.g. 2010) for both your model simulations and the global study. I would also suggest to state mass losses instead of volume losses for easier comparison with other studies (e.g. Kraaijenbrink et al., 2017; Rounce et al., 2023). The 39% difference in projected mass loss for Khumbu Glacier under RCP4.5 between this study and Rounce et al. (2023) is substantial and merits a more thorough discussion. What are the key drivers of this discrepancy - differences in SMB model physics, debris representation, regional precipitation trends, or calibration strategies?
Additional references that might be of interest for the introduction and discussion
Arndt & Schneider (2023): https://doi.org/10.1017/jog.2023.46 => Modelling glacier mass balance and investigating sensitivity to atmospheric forcing in High Mountain Asia
Collier et al. (2013): https://doi.org/10.5194/tc-7-779-2013 => Interactive modelling of glacier-atmosphere interactions
Collier et al. (2015): https://doi.org/10.5194/tc-9-1617-2015 => Modelling the impact of debris cover on ablation and glacier-atmosphere feedbacks
Compagno et al. (2022): https://doi.org/10.5194/tc-16-1697-2022 => Modelling supraglacial debris-cover evolution
Temme et al. (2023): https://doi.org/10.5194/tc-17-2343-2023 => Strategies for SMB model calibration (including COSIPY)
Zekollari et al. (2017): https://doi.org/10.5194/tc-11-805-2017 => Complete different setting, but insightful simulations and discussion on the potential mitigation impact of increasing precipitation under different warming scenarios (see e.g. Fig. 11)
Citation: https://doi.org/10.5194/egusphere-2025-1211-RC2 - AC1: 'Reply on RC1 and RC2', Ann Rowan, 16 Jun 2025
Peer review completion
Post-review adjustments
Journal article(s) based on this preprint
Viewed
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 5,394 | 0 | 15 | 5,409 | 0 | 0 |
- HTML: 5,394
- PDF: 0
- XML: 15
- Total: 5,409
- BibTeX: 0
- EndNote: 0
Viewed (geographical distribution)
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Anya Schlich-Davies
Andrew Ross
Duncan Quincey
Vivi Pedersen
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.