the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Jun 2026
Two-Dimensional Hydrodynamic Modelling of Glacial Lake Outburst Flood Scenarios from Chubda Tsho: Breach Sensitivity and Downstream Flood Dynamics in the Chamkhar Chhu Basin, Bhutan
Abstract. Glacial lake outburst floods (GLOFs) are high-magnitude, short-duration hazards in glacier-fed Himalayan basins, where steep valley confinement can amplify downstream flood intensity. Despite the identification of Chubda Tsho as a potentially dangerous glacial lake in Bhutan, hydrodynamic assessments of potential outburst floods from this lake have not yet been reported. This study evaluates potential GLOF scenarios from Chubda Tsho using inflows generated by a calibrated and independently validated Hydrologic Engineering Center’s Hydrologic Modeling System (HEC-HMS), coupled with two-dimensional (2D) simulations in the Hydrologic Engineering Center's River Analysis System (HEC-RAS). The model was calibrated against the July 2007 flood (NSE = 0.90) and validated using the August 2015 event (NSE = 0.56). Hydrodynamic performance was assessed using an extreme-event hydrograph from Cyclone Aila (2009) and compared with observed inundation extent. Results show good agreement, with a Critical Success Index (CSI) of 0.67, a total area difference of 2.56%, and close correspondence between simulated (1.60 km²) and observed (1.56 km²) inundation extents within the 16.14 km² Chamkhar Valley model domain. The Chamkhar Valley contains approximately 8.41 km² of habitable area. Three breach scenarios representing 50%, 75%, and 100% lake-volume release were simulated using empirically derived breach relationships. Flood waves reach the Chamkhar Valley between approximately 1 h 50 min and 2 h 15 min after breach initiation. Peak discharge increases from 2,114 m³ s⁻¹ to 5,258 m³ s⁻¹, accompanied by nonlinear increases in peak depth (8.4–12.8 m) and velocity (5.0–7.6 m s⁻¹). While inundation extent remains constrained by valley geometry, flood depth and velocity increase disproportionately with breach magnitude. Results indicate that downstream hydraulic intensity is strongly influenced by breach magnitude, while warning time remains limited in confined Himalayan valleys. This highlights the value of validated 2D hydrodynamic modelling for GLOF hazard assessment in Himalayan catchments with limited observational data.
- Preprint
(2442 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2026-2584', Anonymous Referee #1, 07 Jun 2026
-
AC1: 'Reply on RC1', Tenzin Namgay, 27 Aug 2026
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript and for recognising the importance of GLOF hazard assessment in Himalayan catchments. We greatly appreciate the reviewer’s thoughtful comments regarding the scientific contribution, modelling framework, data sources, uncertainty assessment, and comparison with previous GLOF studies. We have carefully considered all of the comments and suggestions and have revised the manuscript accordingly. Our detailed responses to each comment are provided below.
RC1-1 Response:
We sincerely thank the reviewer for this important comment regarding the scientific scope and contribution of the study. We agree that the scientific questions underlying the modelling should be clearly distinguished from a purely site-specific technical hazard assessment.
- The primary scientific focus of this study is to investigate how breach magnitude influences downstream flood-wave dynamics and how these responses are affected by the steep and laterally confined valley geometry of the Chamkhar Chhu system. Three hypothetical scenarios representing 50%, 75%, and 100% release of the Chubda Tsho lake volume were therefore simulated as a breach-sensitivity analysis. The resulting changes in peak discharge, flood depth, flow velocity, arrival time, and flood-wave attenuation were systematically examined. The results show increasing hydraulic intensity with increasing breach magnitude, while lateral flood expansion remains constrained by the valley geometry.
- The study also comprises two separate HEC-RAS model applications with different upstream boundary conditions and purposes. First, the downstream Chamkhar Valley hydraulic setup was evaluated using the Cyclone Aila (2009) rainfall-driven flood, for which the observed inundation extent was obtained from the National Center for Hydrology and Meteorology (NCHM), Bhutan. This simulation was used to evaluate the downstream hydraulic representation, including the adopted Manning’s roughness coefficients, through comparison between simulated and observed inundation. The Cyclone Aila simulation reproduced the observed inundation with a Critical Success Index (CSI) of 0.67 and an inundation-area difference of 2.56%. Importantly, this simulation was not used to calibrate or validate the Chubda Tsho GLOF scenarios.
- Second, a separate GLOF modelling application was conducted by changing the upstream boundary condition to the Chubda Tsho lake outlet and applying synthetic breach hydrographs representing the 50%, 75%, and 100% lake-volume release scenarios. Since no documented GLOF has occurred at Chubda Tsho and corresponding observed GLOF discharge and inundation data are unavailable, direct calibration or validation of the GLOF scenarios is not possible. The three scenarios were therefore designed as sensitivity scenarios to investigate the hydraulic response to different breach magnitudes rather than as validation events.
- Accordingly, the scientific contribution of the study is centred on (1) quantifying the sensitivity of downstream hydraulic response to breach magnitude and (2) examining the influence of steep and confined Himalayan valley geometry on flood-wave propagation and attenuation, while the Cyclone Aila simulation provides an independent evaluation of the downstream hydraulic setup. The reviewer’s suggestions regarding model applicability, hydrological data, and downstream exposure were carefully considered, while the present manuscript maintains its primary focus on breach sensitivity and downstream GLOF dynamics.
We sincerely appreciate the reviewer’s comment, which has helped us clarify the scientific questions, scope, and broader contribution of the study.
RC1-2 Response:
We sincerely thank the reviewer for raising these important points and for providing the opportunity to clarify the modelling framework, data sources, and terrain preparation.
- Regarding sediment entrainment and erosion, we acknowledge that the HEC-RAS framework used in this study does not explicitly simulate progressive moraine erosion, sediment entrainment, or debris transport during GLOF propagation. Representing these processes would require detailed information on moraine and channel-material properties, sediment availability, erodibility, and channel geometry along the complete routing path, which is not available for the study area. The high-altitude upstream reaches are also difficult to access and lack surveyed cross-sections, while available field-survey information is concentrated in the downstream area. Therefore, the present study focuses on water-dominated flood-wave propagation using prescribed breach hydrographs, rather than attempting to reproduce the coupled breach–erosion–sediment transport process. These limitations are acknowledged in the revised manuscript, and incorporation of erosion, sediment entrainment, and debris-flow processes is identified as an important direction for future work when adequate field data become available.
- Regarding the discharge data, we clarify that the observed discharge records were obtained from the National Center for Hydrology and Meteorology (NCHM), Bhutan, and were not generated by HEC-HMS. Observed rainfall and discharge records were used for calibration and independent validation of the HEC-HMS rainfall–runoff model. Thus, the discharge observations were not treated as modelled values.
- Regarding the hydrometeorological datasets, rainfall and discharge observations supplied by NCHM were used for HEC-HMS calibration and validation. The discharge observations were obtained from the Chamkhar Chhu gauging station (Chamkhar; 90.754653° E, 27.549856° N). The HEC-HMS model was calibrated using the July 2007 flood and independently validated using the August 2015 flood, achieving NSE values of 0.90 and 0.56, respectively. The calibrated HEC-HMS model was subsequently used to generate the rainfall–runoff hydrograph for the Cyclone Aila hydraulic evaluation.
- The Cyclone Aila hydraulic evaluation represents a separate HEC-RAS application. The observed Aila inundation extent was obtained from NCHM and was used to evaluate the downstream hydraulic representation. The HEC-HMS-derived Aila rainfall–runoff hydrograph was applied as the upstream boundary condition, and the simulated inundation was compared with the NCHM-observed inundation extent. The model reproduced the observed flood extent with a Critical Success Index (CSI) of 0.67 and an inundation-area difference of 2.56%. This evaluation was used to assess the downstream hydraulic representation and adopted Manning’s roughness coefficients.
- The Chubda Tsho GLOF simulations represent a separate HEC-RAS application. For these simulations, the upstream boundary condition was changed to the Chubda Tsho lake outlet, and synthetic breach hydrographs representing 50%, 75%, and 100% lake-volume release were applied. Therefore, the Cyclone Aila simulation was not used to calibrate or validate the hypothetical Chubda Tsho GLOF scenarios. Since no documented GLOF has occurred at Chubda Tsho and corresponding observed GLOF discharge and inundation data are unavailable, the three cases were treated as breach-sensitivity scenarios rather than validated GLOF events.
- Regarding the terrain data, the ALOS PALSAR DEM (12.5 m) was used as the primary terrain dataset for hydraulic modelling because its finer spatial resolution was considered more suitable for representing the river corridor and surrounding valley terrain. The SRTM DEM (30 m) was used only to fill void areas in the ALOS PALSAR dataset where required and was not used as an alternative hydraulic terrain surface. The resulting ALOS PALSAR terrain was conditioned to improve channel representation and hydraulic connectivity, including channel carving where required and incorporation of available surveyed channel information in the downstream area. The same conditioned terrain was used for both the Cyclone Aila hydraulic evaluation and the Chubda Tsho GLOF simulations.
- A formal quantitative DEM sensitivity analysis was not undertaken because SRTM was used only as a supplementary source for void filling rather than as an alternative hydraulic terrain. Nevertheless, we acknowledge that residual terrain uncertainty remains, particularly in the unsurveyed high-altitude upstream reaches where direct field verification was not possible.
We appreciate the reviewer’s comment, which has helped us clarify the distinction between observed hydrological data, model-derived rainfall–runoff hydrographs, the independent Cyclone Aila hydraulic evaluation, and the separate Chubda Tsho GLOF scenario simulations, as well as the respective roles of the ALOS PALSAR and SRTM DEM datasets.
RC1-3 Response:
We sincerely thank the reviewer for highlighting the importance of uncertainty in GLOF hazard modelling. We agree that uncertainties can arise at different stages of the modelling chain, including hydrometeorological observations, terrain representation, lake characteristics, breach parameterisation, and the assumptions used in hydraulic modelling. We have therefore clarified these sources of uncertainty and their implications for the present results.
- Regarding precipitation and hydrological observations, direct rainfall and discharge observations are not available at or immediately around Chubda Tsho. The available NCHM rainfall and discharge observations were obtained from the relevant monitoring stations. The discharge observations used in this study were obtained from the Chamkhar Chhu gauging station (Chamkhar; 90.754653° E, 27.549856° N), located approximately 47 km downstream of Chubda Tsho, and were used for calibration and validation of the rainfall–runoff model for the gauged catchment. These observations therefore provide catchment-scale hydrological information rather than direct measurements of GLOF discharge at Chubda Tsho. Importantly, precipitation is not the forcing mechanism for the Chubda Tsho GLOF simulations, which use prescribed breach hydrographs as the upstream forcing. The lack of site-specific hydrometeorological observations near Chubda Tsho is therefore recognised as an important source of uncertainty.
- Regarding lake characteristics, the available information is based on the NCHM-reported lake area, storage volume, and maximum depth. The lake area corresponds to the 2016 inventory derived from Sentinel-2 imagery, while the storage volume and maximum depth were obtained from the bathymetric survey conducted in 2010 under the DGM–JICA/JST GLOF Project (2009–2012). Differences in observation periods and the absence of repeated bathymetric surveys introduce uncertainty in the representation of present-day lake storage and geometry. However, because no uncertainty bounds for these parameters were available, they were retained as the reference values for the scenario analysis.
- For the breach parameters and hydrograph, detailed bathymetric, moraine-dam, geotechnical, sediment, and observed breach-process data were not available. Consequently, breach characteristics could not be derived from an observed breach event. Instead, Froehlich (1995, 2008) empirical relationships were used to estimate breach width, formation time, and peak discharge from the available lake-storage and breach-depth information. The resulting hydrographs were constructed using the estimated breach parameters and specified release volumes, with the recession limb adjusted iteratively to conserve the prescribed outburst volume. These empirical relationships provide physically informed, order-of-magnitude estimates under data-scarce conditions but introduce uncertainty because they are not based on site-specific observations of a Chubda Tsho breach.
- To examine the sensitivity of downstream hydraulic response to the uncertain magnitude of lake release, three scenarios representing 50%, 75%, and 100% of the available lake storage were considered. These scenarios are not intended to represent probabilities or deterministic predictions of specific breach events. Rather, they provide a structured sensitivity assessment of how different assumed release magnitudes affect peak discharge, flood depth, flow velocity, arrival time, and inundation characteristics. The resulting differences therefore indicate the relative hydraulic response under different breach-volume assumptions.
- Uncertainty also exists in the terrain representation. The hydraulic modelling uses the conditioned ALOS PALSAR DEM (12.5 m) as the primary terrain dataset, while SRTM (30 m) was used only to fill void areas where required. Although terrain conditioning was undertaken to improve channel representation and hydraulic connectivity, limitations remain in representing complex high-altitude terrain, particularly in upstream areas where field-survey information is unavailable. The lack of surveyed cross-sections in these remote reaches further limits direct verification of the terrain representation.
- We acknowledge that the present study does not provide a formal probabilistic uncertainty propagation or ensemble analysis incorporating all sources of uncertainty simultaneously. Such an analysis would require defensible probability distributions or uncertainty bounds for parameters including lake storage, lake geometry, breach dimensions, breach formation time, hydrograph shape, and terrain errors, which cannot currently be adequately constrained from the available observations. Given these data limitations, we consider the scenario-based breach sensitivity analysis, together with explicit identification of the principal data and modelling uncertainties, to be an appropriate approach for the present study.
Accordingly, the simulated GLOF results should be interpreted as scenario-based estimates of potential downstream hydraulic response rather than deterministic predictions of a future GLOF. We appreciate the reviewer’s comment, which has helped us improve the transparency of the uncertainty sources, their roles within the modelling framework, and their implications for interpretation of the results.
RC1-4 Response
We sincerely thank the reviewer for this valuable suggestion. We agree that a broader literature context is important for establishing the scientific relevance of the study beyond the Chubda Tsho study area.
- The literature review and discussion have been expanded to include additional studies on GLOF occurrence, hazard assessment, and hydrodynamic modelling in Bhutan and the wider Himalayan region. In particular, the revised discussion places the findings of the present study in relation to previous GLOF investigations from Bhutan and Nepal, with emphasis on flood-wave propagation, breach magnitude, downstream hydraulic response, flood arrival time, and the influence of steep and confined mountain-valley geometry.
- The revised discussion compares the simulated flood response with previous GLOF studies from Bhutan and Nepal, particularly in terms of peak discharge, flood-wave propagation, arrival time, inundation behaviour, and the influence of valley confinement. The comparison highlights both similarities and differences between the present results and previous studies and identifies the implications of these findings for understanding GLOF dynamics in confined Himalayan valleys. This comparison helps demonstrate that the study contributes not only a site-specific assessment of Chubda Tsho, but also a scenario-based evaluation of how breach magnitude and valley geometry influence downstream GLOF dynamics in a data-scarce Himalayan setting.
- We thank the reviewer for this recommendation, which has helped us strengthen the literature context, expand the discussion of previous GLOF studies from Bhutan and Nepal, and more clearly articulate the broader scientific contribution of the study.
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript. The comments have helped us clarify the scientific questions and contribution of the study, strengthen the description of the modelling framework and data sources, more explicitly address uncertainties and modelling limitations, and expand the discussion and comparison with previous GLOF studies in Bhutan and the wider Himalayan region. We believe that these revisions have substantially improved the clarity, transparency, scientific interpretation, and overall quality of the manuscript. We are grateful to the reviewer for the valuable suggestions and constructive guidance, which have helped us strengthen the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-2584-AC1
-
AC1: 'Reply on RC1', Tenzin Namgay, 27 Aug 2026
-
RC2: 'Comment on egusphere-2026-2584', Anonymous Referee #2, 24 Jun 2026
Summary
The study by Namgay et al. (2026) models three different glacial lake outburst flood (GLOF) scenarios originating from Chubda Tsho, located in the headwaters of the Chamkar Chu basin, using HEC-HMS and HEC-RAS. While the topic is highly relevant to local communities and disaster risk management in Bhutan, the manuscript currently offers limited scientific novelty in terms of methodology, analysis, and contribution to the existing body of knowledge on GLOF hazard and risk assessments. I encourage the authors to better articulate the novelty and significance of their work and to address the following comments before the manuscript can be considered for publication.
Major Comments
1. Chubda Tsho glacial lake characterization
The manuscript provides insufficient information on Chubda Tsho itself. The authors should include details on:
- The origin and formation history of the lake.
- Lake type (e.g., moraine-dammed, ice-contact, etc.).
- Historical lake expansion and development rates.
- The basis on which the lake was identified as a potentially dangerous glacial lake.
- Current lake area and volume, including methods used to derive these estimates.
- The area–volume relationship applied in the study.
- The reference year(s) used for lake area and volume estimates.
- Potential future lake expansion and corresponding volume increases.
- The likely maximum extent of future lake growth.
- In addition, more information on the parent glacier (Chubda Glacier) should be provided, including available estimates of glacier mass balance, velocity, thickness, and recent changes derived from remote sensing studies.
2. Flood and rainfall events used for calibration and validation
The manuscript provides insufficient information regarding the flood events used in the modelling framework (e.g., the 2007 flood, the 2015 event, and Cyclone Aila). The authors should clearly describe:
- The characteristics of each event.
- The hydrometeorological parameters used.
- Data sources and the agencies or organizations responsible for data collection.
- The rationale for selecting certain events for calibration and others for validation.
A more transparent explanation of the calibration and validation strategy is necessary.
3. Hydrograph assumptions and model validation
A fundamental difference exists between rainfall-induced flood hydrographs and GLOF hydrographs. The manuscript states that: “Hydrodynamic validation relied on the extreme rainfall-driven flood associated with Cyclone Aila (2009), in the absence of documented historical GLOF observations for the study basin.” This approach requires stronger justification because rainfall floods and moraine-dam breach floods are governed by different hydrological and hydraulic processes and often exhibit substantially different hydrograph shapes. I recommend validating the modelling framework against the well-documented 1994 Lugge Tsho GLOF in Bhutan, for which inundation extent and other observational datasets are available. Such validation would provide greater confidence in the model's ability to simulate GLOF dynamics.
Furthermore, the manuscript assumes that: “Synthetic outburst hydrographs were constructed assuming a single-peak breach process, with peak discharge occurring near the midpoint of breach development.” The authors should provide stronger justification for this assumption. Moraine-dam breach hydrographs are often characterized by a very steep rising limb and rapid peak discharge development. The current assumptions may not adequately capture the hydrograph characteristics of a moraine-dam failure. A comparison with documented GLOF hydrographs from similar moraine-dammed lakes would strengthen the analysis.
4. Downstream risk assessment
The study focuses primarily on flood propagation but provides limited assessment of downstream impacts. I recommend conducting a more comprehensive downstream risk assessment, similar to the approach adopted by Rinzin et al. (2023), including:
- Exposure of settlements and infrastructure.
- Potential impacts on hydropower facilities, roads, bridges, and agricultural land.
- Population at risk.
- Hazard and risk zonation.
Such an analysis would significantly enhance the practical relevance of the study.
5. Present-day and future GLOF scenarios
The manuscript would be substantially strengthened by incorporating future GLOF scenarios in addition to the present-day assessment. Given the ongoing expansion of Chubda Tsho, the authors should evaluate how future increases in lake area and volume may influence potential outburst magnitude, peak discharge, inundation extent, and downstream impacts. This could be achieved by projecting plausible future lake extents and volumes based on observed lake growth rates, glacier retreat, and topographic constraints.
Importantly, such an analysis could provide a clear element of scientific novelty for the study. While many existing GLOF assessments focus solely on present-day conditions, a comparative evaluation of present and future GLOF hazards would offer valuable insights into how climate-driven lake evolution may alter downstream flood risk over time. This would significantly enhance the study's relevance for long-term disaster risk reduction, infrastructure planning, and climate adaptation in the Chamkar Chu basin.
By explicitly comparing present-day and future hazard and risk scenarios, the authors would move beyond a conventional GLOF modelling exercise and provide a more forward-looking assessment of evolving cryospheric hazards.
6. Figure quality and visualization
All figures are currently below publication quality and require substantial improvement. In particular, Figures 3–7 and Figures 10–12 should be revised to improve:
- Resolution and readability.
- Map layout and cartographic quality.
- Color schemes and contrast.
- Label clarity and font size.
- Figure captions and explanatory content.
High-quality visualizations are essential for effectively communicating the modelling results and their implications.
Overall Recommendation
The study addresses an important hazard issue for the Chamkar Chu basin and downstream communities. However, the manuscript would be substantially strengthened by providing a more comprehensive characterization of Chubda Tsho and its parent glacier, improving the justification and validation of the modelling framework, incorporating future lake development and GLOF scenarios, expanding the downstream risk assessment, and enhancing figure quality. Most importantly, the authors should clearly demonstrate the scientific novelty and added value of the study relative to existing GLOF modelling research.
Rinzin, S., Zhang, G., Sattar, A., Wangchuk, S., Allen, S. K., Dunning, S., and Peng, M.: GLOF hazard, exposure, vulnerability, and risk assessment of potentially dangerous glacial lakes in the Bhutan Himalaya, J. Hydrol., 619, 129311, https://doi.org/10.1016/j.jhydrol.2023.129311, 2023.
Citation: https://doi.org/10.5194/egusphere-2026-2584-RC2 -
AC2: 'Reply on RC2', Tenzin Namgay, 27 Aug 2026
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript and for the valuable suggestions provided to strengthen the study. We appreciate the reviewer’s comments regarding the characterization of Chubda Tsho and its parent glacier, the hydrological and hydrodynamic modelling framework, validation strategy, uncertainty and hydrograph assumptions, downstream implications, future lake evolution, and figure quality. We have carefully considered all of the comments and suggestions and have revised the manuscript accordingly. Detailed responses to each comment are provided below.
RC2-1 Response
We sincerely thank the reviewer for highlighting the need for a more comprehensive characterization of Chubda Tsho and its parent glacier. We have revised the manuscript to provide clearer information on the lake characteristics, historical development, data sources, and limitations of the available observations.
- Chubda Tsho (Cham_gl 383 in the NCHM inventory) is located in the headwaters of the Chamkhar Chhu basin at approximately 4,868 m a.s.l. The lake is associated with an ice-cored moraine and a relatively flat threshold at the end moraine and is directly fed by Cham_gr 71. Earlier investigations reported dead ice beneath the moraine, active slides, and fresh ice cliffs around the lake and recommended continued monitoring and detailed investigation. These observations indicate a moraine-dammed glacial lake with evidence of ongoing glacier–moraine interaction and historical lake development. Based on these geomorphological and dam-related characteristics, Chubda Tsho was identified and retained as a potentially dangerous glacial lake (PDGL) in the NCHM assessment.
- Regarding historical lake development, NCHM (2019) reports a surface area of approximately 1.035 km² in 2001 and 1.388 km² in 2016, with the latter derived from Sentinel-2 imagery. The report also cites an earlier expansion rate of approximately 0.027 km² yr⁻¹ and indicates an approximately 564 m upward expansion between 2001 and 2016. These values represent historical observations and are not treated as direct projections of future lake growth.
- The lake area, storage, and maximum depth used in this study were obtained from NCHM (2019). The reported lake area is approximately 1.39 km² and corresponds to the 2016 NCHM inventory derived from Sentinel-2 imagery. The reported storage volume of 21.69 × 10⁶ m³ and maximum depth of 56 m were obtained from the bathymetric survey conducted in 2010 under the DGM–JICA/JST GLOF Project (2009–2012). No separate area–volume relationship was applied, as the storage estimate was already available from the bathymetric assessment. Differences in observation periods and the absence of repeated bathymetric surveys introduce uncertainty in the representation of present-day lake storage and geometry.
- The corrected DEM was also used to characterize the lake and outlet elevations, while the NCHM-reported lake storage and maximum depth were retained as the basis for defining the GLOF scenarios.
- Regarding future lake expansion, although historical expansion has been documented, the available observations do not provide a sufficiently constrained basis for projecting future lake area, volume, or maximum future extent. In particular, the available data do not provide sufficient information to establish a physically defensible future lake-growth trajectory or a reliable limiting lake extent. We therefore do not introduce an unsupported future-growth projection into the present simulations.
- For the parent glacier, NCHM identifies Cham_gr 71 as the glacier directly feeding Chubda Tsho. However, detailed site-specific observations of glacier mass balance, ice thickness, surface velocity, and recent glacier change were not available in the datasets used in this study. These parameters were therefore not incorporated into the hydrodynamic modelling and are acknowledged as data limitations and priorities for future investigation.
- Because site-specific breach observations and detailed geotechnical information for the moraine were unavailable, the potential outburst magnitude was represented using 50%, 75%, and 100% lake-storage release scenarios, corresponding to 10.845 × 10⁶, 16.2675 × 10⁶, and 21.69 × 10⁶ m³, respectively. Breach width, formation time, and peak discharge were estimated using the Froehlich (1995, 2008) empirical relationships, and the resulting hydrographs were constructed to conserve the specified outburst volumes.
We have revised the manuscript to clearly distinguish between historical observations, NCHM-derived lake parameters, modelling assumptions, and limitations associated with future lake evolution and glacier dynamics. We appreciate the reviewer’s comment, which has helped us improve the transparency and completeness of the Chubda Tsho characterization.
RC2-2 Response
We sincerely thank the reviewer for this comment. We agree that the roles of the different flood events and the calibration–validation strategy were not sufficiently clear in the original manuscript. We have revised the manuscript to clarify the characteristics, data sources, and distinct purposes of the 2007, 2015, and Cyclone Aila (2009) events.
- For the HEC-HMS hydrological model, observed rainfall and discharge data obtained from the National Center for Hydrology and Meteorology (NCHM), Bhutan, were used. The July 2007 flood event was selected for model calibration, while the August 2015 flood event was used for independent validation. These events were selected based on the availability of corresponding observed rainfall and discharge records, allowing the rainfall–runoff model to be calibrated and subsequently evaluated using an independent event. The discharge observations were obtained from the Chamkhar Chhu gauging station (Chamkhar; 90.754653° E, 27.549856° N), located approximately 47 km downstream of Chubda Tsho. The HEC-HMS model achieved NSE values of 0.90 for the 2007 calibration event and 0.56 for the 2015 validation event.
- The Cyclone Aila (2009) event was treated separately from the HEC-HMS calibration and validation events. It was selected specifically to evaluate the downstream HEC-RAS hydraulic setup, because an observed inundation extent for the Cyclone Aila flood was available in the downstream Chamkhar area from NCHM. The calibrated HEC-HMS model was used to generate the rainfall–runoff hydrograph for Aila, which was then applied as the upstream boundary condition of the downstream HEC-RAS model. The simulated inundation was compared with the observed Aila inundation extent, resulting in a CSI of 0.67 and an inundation-area difference of 2.56%. This evaluation was used to assess the downstream hydraulic representation and adopted Manning’s roughness coefficients.
- Importantly, the Cyclone Aila simulation was not used to calibrate or validate the Chubda Tsho GLOF simulations. The two applications have different purposes and upstream boundary conditions. For Cyclone Aila, the rainfall-driven hydrograph was applied to the downstream Chamkhar modelling domain. For the GLOF simulations, the upstream boundary condition was changed to the Chubda Tsho lake outlet, and synthetic breach hydrographs representing 50%, 75%, and 100% lake-storage release scenarios were applied.
- Because no documented historical GLOF has occurred at Chubda Tsho and corresponding observed GLOF discharge and inundation data are unavailable, direct calibration or validation of the Chubda Tsho GLOF simulations is not possible. The three GLOF scenarios were therefore designed as breach-sensitivity scenarios to systematically examine the downstream hydraulic response to different assumed release magnitudes.
The revised manuscript now provides greater detail on the event periods, observed rainfall and discharge data, hydrological and hydrodynamic model parameters, boundary conditions, and data sources, while clearly distinguishing the 2007 HEC-HMS calibration, 2015 HEC-HMS validation, 2009 Cyclone Aila downstream HEC-RAS evaluation, and separate Chubda Tsho GLOF scenario simulations. We appreciate the reviewer’s comment, which has helped us improve the transparency of the modelling and validation strategy.
RC2-3 Response:
We sincerely thank the reviewer for this important comment. We agree that rainfall-induced floods and moraine-dam breach floods are governed by different processes and can have substantially different hydrograph characteristics. We have therefore clarified the purpose of the Cyclone Aila assessment and the assumptions associated with the synthetic GLOF hydrographs.
- First, Cyclone Aila (2009) was not used to validate the GLOF breach process itself. It was used to independently evaluate the downstream HEC-RAS hydraulic setup under an observed extreme-flood condition, because an observed inundation extent was available for the downstream Chamkhar area. The Aila rainfall–runoff hydrograph generated using the calibrated HEC-HMS model was applied as the upstream boundary condition of the downstream HEC-RAS model, and the simulated inundation was compared with the observed Aila inundation extent. The simulation achieved a CSI of 0.67 and an inundation-area difference of 2.56%. This assessment was therefore intended to evaluate the representation of downstream hydraulic propagation, terrain, channel connectivity, and roughness, rather than to demonstrate equivalence between rainfall-flood and GLOF hydrographs.
- We appreciate the reviewer’s suggestion to use the 1994 Lugge Tsho GLOF as an additional reference. Although the 1994 Lugge Tsho event provides a valuable regional reference, Lugge Tsho is located in Punakha District, whereas Chubda Tsho is located in the Chamkhar Chhu basin of Bumthang District, and the two events occurred in different lake–river systems. Direct use of the Lugge Tsho event for validation of the present Chubda Tsho model would therefore require transferring observations between systems with differences in valley and river-channel geometry, terrain, LULC and associated Manning’s roughness, river perimeter, and model-domain configuration. The hydraulic model in this study was specifically developed for the Chubda Tsho–Chamkhar Chhu system. For this reason, we did not use the Lugge Tsho event as a direct validation case. Instead, the simulated GLOF responses were compared with reported results from documented Himalayan GLOF studies, including studies from Bhutan, to assess whether the magnitude and behaviour of the simulated responses are within physically plausible ranges. This comparison is treated as a plausibility assessment rather than formal validation.
- Regarding the synthetic GLOF hydrographs, no observed breach hydrograph or site-specific breach observations are available for Chubda Tsho. The breach characteristics were therefore estimated using the Froehlich (1995, 2008) empirical relationships from the available lake-storage and breach-depth information. For the 50%, 75%, and 100% release scenarios, the estimated breach formation times were approximately 0.68, 0.58, and 0.52 h, with corresponding estimated peak discharges of approximately 4,498, 8,382, and 13,035 m³ s⁻¹.
- The resulting hydrographs were constructed using an idealised two-part, single-peak formulation, consisting of a linear rising limb followed by an exponential recession limb. The peak was placed near the midpoint of the estimated breach-formation period as a simplified representation of the timing of peak discharge during breach development in the absence of observations defining the actual temporal evolution of the breach. The recession constant was then adjusted iteratively so that the integrated hydrograph volume matched the prescribed released lake volume for each scenario. The resulting final peak discharges were 4,412, 8,189, and 12,782 m³ s⁻¹ for the 50%, 75%, and 100% scenarios, respectively.
- We acknowledge that this idealised formulation cannot reproduce all characteristics of an actual moraine-dam failure, including potentially very rapid peak development, progressive breach enlargement, sediment entrainment, and erosion–deposition feedbacks. These processes could substantially influence the shape and magnitude of a real GLOF hydrograph. However, the necessary site-specific breach, sediment, and erosion observations are unavailable for Chubda Tsho. The synthetic hydrographs are therefore used consistently as scenario-based forcing conditions to examine the sensitivity of downstream hydraulic response to different assumed release magnitudes, rather than as deterministic predictions of an actual future breach hydrograph.
We have revised the manuscript to explicitly distinguish (i) the Cyclone Aila assessment of the downstream hydraulic model, (ii) the hypothetical Chubda Tsho breach-sensitivity scenarios, and (iii) comparison with published GLOF results as a plausibility check. The assumptions and limitations associated with the synthetic hydrograph formulation have also been clarified in the revised manuscript. We sincerely appreciate the reviewer’s comment, which has helped us improve the transparency and scientific interpretation of the modelling framework.
RC2-4 Response:
We sincerely thank the reviewer for this valuable suggestion. We agree that linking the simulated flood dynamics with downstream exposure is important for demonstrating the practical significance of the modelling results.
- The primary focus of the present manuscript is the hydrodynamic behaviour and breach-magnitude sensitivity of GLOF propagation from Chubda Tsho, including peak discharge, flood depth, flow velocity, arrival time, and inundation characteristics. A comprehensive downstream exposure, economic-loss, vulnerability, and evacuation assessment involves additional socioeconomic and asset-specific analyses and is therefore treated as a complementary component of the broader research.
- Nevertheless, in response to the reviewer’s suggestion, we have strengthened the revised manuscript by explicitly discussing the potential exposure of downstream settlements and critical infrastructure within the modelled floodplain, including roads, bridges, agricultural areas, and other major infrastructure. The spatial implications of the simulated inundation for the downstream Chamkhar area are also discussed in relation to the simulated hazard patterns.
- A quantitative population-at-risk assessment was not undertaken in the present manuscript because sufficiently detailed spatially disaggregated population data were not available. Similarly, detailed economic-loss and vulnerability estimates require asset-specific valuation and vulnerability information that could not be reliably constrained using the available datasets. We have therefore avoided presenting potentially uncertain estimates as definitive risk values.
- The revised manuscript now more clearly distinguishes between the hydrodynamic hazard assessment presented here and complementary downstream exposure and risk analyses. This allows the present study to maintain its primary focus on hydrodynamic behaviour and breach-magnitude sensitivity while providing appropriate discussion of the practical implications of the simulated GLOF hazard.
The revised manuscript now more clearly distinguishes between the hydrodynamic hazard assessment presented here and the complementary downstream exposure and risk analyses. We appreciate the reviewer’s recommendation, which has helped us strengthen the discussion of the practical implications of the simulated GLOF hazard while maintaining the scientific focus of the manuscript.
RC2-5 Response:
We sincerely thank the reviewer for this valuable suggestion. We agree that future lake evolution is an important consideration for long-term GLOF hazard assessment, particularly given the documented historical expansion of Chubda Tsho.
- As clarified in the revised manuscript, historical lake expansion has been documented, with the lake area increasing from approximately 1.035 km² in 2001 to 1.388 km² in 2016. However, the available observations do not provide a sufficiently constrained basis for reliably projecting future lake area, storage volume, or maximum lake extent. In particular, consistent multi-temporal bathymetric information and site-specific data on glacier mass balance, ice thickness, and glacier dynamics are not available.
- We therefore considered it inappropriate to extrapolate the historical expansion rate directly to construct future lake scenarios, as this would introduce additional unsupported assumptions into the breach and hydrodynamic modelling. Instead, the present study focuses on scenario-based sensitivity to different outburst magnitudes using the currently available lake storage, represented by the 50%, 75%, and 100% release scenarios. These scenarios are intended to examine the sensitivity of downstream flood dynamics to increasing release magnitude and are not intended to represent future climate scenarios or probabilistic predictions.
- We have clarified this distinction in the revised manuscript and explicitly acknowledged future lake expansion and glacier evolution as important uncertainties and priorities for future research. A physically constrained assessment of future GLOF hazards would require additional multi-temporal lake and glacier observations, repeated bathymetric surveys, and coupled glacier–lake evolution modelling.
We sincerely appreciate the reviewer’s recommendation, which has helped us clarify the scope and limitations of the present study and identify future lake evolution as an important direction for extending GLOF hazard assessment in the Chamkhar Chhu basin.
RC2-6 Response:
We sincerely thank the reviewer for this helpful comment. We have carefully revised the figures throughout the manuscript to improve their resolution, readability, cartographic layout, colour contrast, label visibility, font sizes, and overall presentation quality. The map layouts have also been refined to improve the clarity of spatial information, including appropriate legends, scale bars, north arrows, and coordinate information where applicable. Figure captions have been revised to provide clearer and more informative descriptions of the information presented.
- The revised figures provide clearer visualization of the hydrodynamic model setup, GLOF scenarios, flood propagation, and downstream hydraulic results, with improved consistency in formatting and presentation throughout the manuscript.
- We sincerely appreciate the reviewer’s constructive suggestion, which has helped us improve the overall visual quality, clarity, and readability of the manuscript.
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript. The reviewer’s comments have helped us substantially improve the characterization of Chubda Tsho and its parent glacier, clarify the hydrological and hydrodynamic modelling and validation framework, strengthen the discussion of hydrograph assumptions and uncertainties, better articulate the downstream implications of the simulated GLOF hazard, clarify the scope and limitations of the present-day scenario analysis, and substantially improve the quality and presentation of the figures. We believe that these revisions have improved the clarity, transparency, scientific interpretation, and overall quality of the manuscript, while maintaining its primary focus on breach-magnitude sensitivity and downstream GLOF dynamics in the Chubda Tsho–Chamkhar Chhu system. We are sincerely grateful to the reviewer for the valuable recommendations and constructive guidance, which have helped us strengthen the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-2584-AC2
Status: closed
-
RC1: 'Comment on egusphere-2026-2584', Anonymous Referee #1, 07 Jun 2026
Review egusphere-2026-2584
The authors have chosen one of the lakes identified as potentially hazardous in Bhutan for a hydrodynamic simulation to assess flood extents under a variety of breach scenarios. They use the HEC modelling suite for both the estimated discharge peaks from a hydrological model (if I understand correctly) as well as the hydrodynamical modelling and come to the general conclusion that flood waves in the Chamkar Valley are rapid, leaving little time for early warning. In general, such studies are interesting and required to get a better understanding of potential risks and general preparedness. The manuscript is also very well written, and the figures are generally clear, hence leaving hardly anything for me to comment on in detail. However, there are four crucial shortcomings, that lead me to suggest a thorough rethinking of the scope before resubmitting.
- In general, hazard modelling studies are of interest, also for this specific journal, however to warrant a scientific publication, I believe there need to be more scientific questions associated to the problem. As it stands, the model is applied for one lake and lead times and extents are presented, which could also be done in a technical report. I would expect a specific challenge that is addressed here, whether related to the suitability of HEC in such a complex domain (by comparing to more detailed model outputs), the suitability of the climate and discharge data to calibrate or the sensitivity of the results to expected breaching mechanisms specific to Chamkkar Chhu. Questions around accurate exposure mapping could also be attempted.
- The authors choose HEC, which in general is ok to use for flood cases, but it does not deal well with entrainment/erosion if not addressed specifically. Additionally, if I understand correctly, the authors model the discharge, which they then call ‘observed’ – which if true is problematic – and use this modelled HEC-HMS data to calibrate and validate the HEC-RAS model. That packs a lot of uncertainty into the results that remain completely unaddressed. It’s also unclear where the data the authors claim to have received from the Hydrological Agency come in, at which location and how they were used. Adding to that, the authors note two types of DEM used (30 m and 12.5 m) but it is unclear, which is used where and why, and how sensitive the results are to this choice.
- In general, the study lacks any uncertainty assessment – from the input data you claim to use but d not present (precipitation), to the DEM, the lake area, the lake volume, the breach curve all parts of the model likely have some uncertainty that would propagate to the final results. For a scientific publication this needs to be thoroughly addressed.
- The literature consulted is rather short – general studies that have collected GLOF events in recent years including in Bhutan are not provided and there is no clear Discussion versus the take home messages vs similar studies in Bhutan and Nepal. This needs to be further expanded to create added value for the scientific community.
Citation: https://doi.org/10.5194/egusphere-2026-2584-RC1 -
AC1: 'Reply on RC1', Tenzin Namgay, 27 Aug 2026
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript and for recognising the importance of GLOF hazard assessment in Himalayan catchments. We greatly appreciate the reviewer’s thoughtful comments regarding the scientific contribution, modelling framework, data sources, uncertainty assessment, and comparison with previous GLOF studies. We have carefully considered all of the comments and suggestions and have revised the manuscript accordingly. Our detailed responses to each comment are provided below.
RC1-1 Response:
We sincerely thank the reviewer for this important comment regarding the scientific scope and contribution of the study. We agree that the scientific questions underlying the modelling should be clearly distinguished from a purely site-specific technical hazard assessment.
- The primary scientific focus of this study is to investigate how breach magnitude influences downstream flood-wave dynamics and how these responses are affected by the steep and laterally confined valley geometry of the Chamkhar Chhu system. Three hypothetical scenarios representing 50%, 75%, and 100% release of the Chubda Tsho lake volume were therefore simulated as a breach-sensitivity analysis. The resulting changes in peak discharge, flood depth, flow velocity, arrival time, and flood-wave attenuation were systematically examined. The results show increasing hydraulic intensity with increasing breach magnitude, while lateral flood expansion remains constrained by the valley geometry.
- The study also comprises two separate HEC-RAS model applications with different upstream boundary conditions and purposes. First, the downstream Chamkhar Valley hydraulic setup was evaluated using the Cyclone Aila (2009) rainfall-driven flood, for which the observed inundation extent was obtained from the National Center for Hydrology and Meteorology (NCHM), Bhutan. This simulation was used to evaluate the downstream hydraulic representation, including the adopted Manning’s roughness coefficients, through comparison between simulated and observed inundation. The Cyclone Aila simulation reproduced the observed inundation with a Critical Success Index (CSI) of 0.67 and an inundation-area difference of 2.56%. Importantly, this simulation was not used to calibrate or validate the Chubda Tsho GLOF scenarios.
- Second, a separate GLOF modelling application was conducted by changing the upstream boundary condition to the Chubda Tsho lake outlet and applying synthetic breach hydrographs representing the 50%, 75%, and 100% lake-volume release scenarios. Since no documented GLOF has occurred at Chubda Tsho and corresponding observed GLOF discharge and inundation data are unavailable, direct calibration or validation of the GLOF scenarios is not possible. The three scenarios were therefore designed as sensitivity scenarios to investigate the hydraulic response to different breach magnitudes rather than as validation events.
- Accordingly, the scientific contribution of the study is centred on (1) quantifying the sensitivity of downstream hydraulic response to breach magnitude and (2) examining the influence of steep and confined Himalayan valley geometry on flood-wave propagation and attenuation, while the Cyclone Aila simulation provides an independent evaluation of the downstream hydraulic setup. The reviewer’s suggestions regarding model applicability, hydrological data, and downstream exposure were carefully considered, while the present manuscript maintains its primary focus on breach sensitivity and downstream GLOF dynamics.
We sincerely appreciate the reviewer’s comment, which has helped us clarify the scientific questions, scope, and broader contribution of the study.
RC1-2 Response:
We sincerely thank the reviewer for raising these important points and for providing the opportunity to clarify the modelling framework, data sources, and terrain preparation.
- Regarding sediment entrainment and erosion, we acknowledge that the HEC-RAS framework used in this study does not explicitly simulate progressive moraine erosion, sediment entrainment, or debris transport during GLOF propagation. Representing these processes would require detailed information on moraine and channel-material properties, sediment availability, erodibility, and channel geometry along the complete routing path, which is not available for the study area. The high-altitude upstream reaches are also difficult to access and lack surveyed cross-sections, while available field-survey information is concentrated in the downstream area. Therefore, the present study focuses on water-dominated flood-wave propagation using prescribed breach hydrographs, rather than attempting to reproduce the coupled breach–erosion–sediment transport process. These limitations are acknowledged in the revised manuscript, and incorporation of erosion, sediment entrainment, and debris-flow processes is identified as an important direction for future work when adequate field data become available.
- Regarding the discharge data, we clarify that the observed discharge records were obtained from the National Center for Hydrology and Meteorology (NCHM), Bhutan, and were not generated by HEC-HMS. Observed rainfall and discharge records were used for calibration and independent validation of the HEC-HMS rainfall–runoff model. Thus, the discharge observations were not treated as modelled values.
- Regarding the hydrometeorological datasets, rainfall and discharge observations supplied by NCHM were used for HEC-HMS calibration and validation. The discharge observations were obtained from the Chamkhar Chhu gauging station (Chamkhar; 90.754653° E, 27.549856° N). The HEC-HMS model was calibrated using the July 2007 flood and independently validated using the August 2015 flood, achieving NSE values of 0.90 and 0.56, respectively. The calibrated HEC-HMS model was subsequently used to generate the rainfall–runoff hydrograph for the Cyclone Aila hydraulic evaluation.
- The Cyclone Aila hydraulic evaluation represents a separate HEC-RAS application. The observed Aila inundation extent was obtained from NCHM and was used to evaluate the downstream hydraulic representation. The HEC-HMS-derived Aila rainfall–runoff hydrograph was applied as the upstream boundary condition, and the simulated inundation was compared with the NCHM-observed inundation extent. The model reproduced the observed flood extent with a Critical Success Index (CSI) of 0.67 and an inundation-area difference of 2.56%. This evaluation was used to assess the downstream hydraulic representation and adopted Manning’s roughness coefficients.
- The Chubda Tsho GLOF simulations represent a separate HEC-RAS application. For these simulations, the upstream boundary condition was changed to the Chubda Tsho lake outlet, and synthetic breach hydrographs representing 50%, 75%, and 100% lake-volume release were applied. Therefore, the Cyclone Aila simulation was not used to calibrate or validate the hypothetical Chubda Tsho GLOF scenarios. Since no documented GLOF has occurred at Chubda Tsho and corresponding observed GLOF discharge and inundation data are unavailable, the three cases were treated as breach-sensitivity scenarios rather than validated GLOF events.
- Regarding the terrain data, the ALOS PALSAR DEM (12.5 m) was used as the primary terrain dataset for hydraulic modelling because its finer spatial resolution was considered more suitable for representing the river corridor and surrounding valley terrain. The SRTM DEM (30 m) was used only to fill void areas in the ALOS PALSAR dataset where required and was not used as an alternative hydraulic terrain surface. The resulting ALOS PALSAR terrain was conditioned to improve channel representation and hydraulic connectivity, including channel carving where required and incorporation of available surveyed channel information in the downstream area. The same conditioned terrain was used for both the Cyclone Aila hydraulic evaluation and the Chubda Tsho GLOF simulations.
- A formal quantitative DEM sensitivity analysis was not undertaken because SRTM was used only as a supplementary source for void filling rather than as an alternative hydraulic terrain. Nevertheless, we acknowledge that residual terrain uncertainty remains, particularly in the unsurveyed high-altitude upstream reaches where direct field verification was not possible.
We appreciate the reviewer’s comment, which has helped us clarify the distinction between observed hydrological data, model-derived rainfall–runoff hydrographs, the independent Cyclone Aila hydraulic evaluation, and the separate Chubda Tsho GLOF scenario simulations, as well as the respective roles of the ALOS PALSAR and SRTM DEM datasets.
RC1-3 Response:
We sincerely thank the reviewer for highlighting the importance of uncertainty in GLOF hazard modelling. We agree that uncertainties can arise at different stages of the modelling chain, including hydrometeorological observations, terrain representation, lake characteristics, breach parameterisation, and the assumptions used in hydraulic modelling. We have therefore clarified these sources of uncertainty and their implications for the present results.
- Regarding precipitation and hydrological observations, direct rainfall and discharge observations are not available at or immediately around Chubda Tsho. The available NCHM rainfall and discharge observations were obtained from the relevant monitoring stations. The discharge observations used in this study were obtained from the Chamkhar Chhu gauging station (Chamkhar; 90.754653° E, 27.549856° N), located approximately 47 km downstream of Chubda Tsho, and were used for calibration and validation of the rainfall–runoff model for the gauged catchment. These observations therefore provide catchment-scale hydrological information rather than direct measurements of GLOF discharge at Chubda Tsho. Importantly, precipitation is not the forcing mechanism for the Chubda Tsho GLOF simulations, which use prescribed breach hydrographs as the upstream forcing. The lack of site-specific hydrometeorological observations near Chubda Tsho is therefore recognised as an important source of uncertainty.
- Regarding lake characteristics, the available information is based on the NCHM-reported lake area, storage volume, and maximum depth. The lake area corresponds to the 2016 inventory derived from Sentinel-2 imagery, while the storage volume and maximum depth were obtained from the bathymetric survey conducted in 2010 under the DGM–JICA/JST GLOF Project (2009–2012). Differences in observation periods and the absence of repeated bathymetric surveys introduce uncertainty in the representation of present-day lake storage and geometry. However, because no uncertainty bounds for these parameters were available, they were retained as the reference values for the scenario analysis.
- For the breach parameters and hydrograph, detailed bathymetric, moraine-dam, geotechnical, sediment, and observed breach-process data were not available. Consequently, breach characteristics could not be derived from an observed breach event. Instead, Froehlich (1995, 2008) empirical relationships were used to estimate breach width, formation time, and peak discharge from the available lake-storage and breach-depth information. The resulting hydrographs were constructed using the estimated breach parameters and specified release volumes, with the recession limb adjusted iteratively to conserve the prescribed outburst volume. These empirical relationships provide physically informed, order-of-magnitude estimates under data-scarce conditions but introduce uncertainty because they are not based on site-specific observations of a Chubda Tsho breach.
- To examine the sensitivity of downstream hydraulic response to the uncertain magnitude of lake release, three scenarios representing 50%, 75%, and 100% of the available lake storage were considered. These scenarios are not intended to represent probabilities or deterministic predictions of specific breach events. Rather, they provide a structured sensitivity assessment of how different assumed release magnitudes affect peak discharge, flood depth, flow velocity, arrival time, and inundation characteristics. The resulting differences therefore indicate the relative hydraulic response under different breach-volume assumptions.
- Uncertainty also exists in the terrain representation. The hydraulic modelling uses the conditioned ALOS PALSAR DEM (12.5 m) as the primary terrain dataset, while SRTM (30 m) was used only to fill void areas where required. Although terrain conditioning was undertaken to improve channel representation and hydraulic connectivity, limitations remain in representing complex high-altitude terrain, particularly in upstream areas where field-survey information is unavailable. The lack of surveyed cross-sections in these remote reaches further limits direct verification of the terrain representation.
- We acknowledge that the present study does not provide a formal probabilistic uncertainty propagation or ensemble analysis incorporating all sources of uncertainty simultaneously. Such an analysis would require defensible probability distributions or uncertainty bounds for parameters including lake storage, lake geometry, breach dimensions, breach formation time, hydrograph shape, and terrain errors, which cannot currently be adequately constrained from the available observations. Given these data limitations, we consider the scenario-based breach sensitivity analysis, together with explicit identification of the principal data and modelling uncertainties, to be an appropriate approach for the present study.
Accordingly, the simulated GLOF results should be interpreted as scenario-based estimates of potential downstream hydraulic response rather than deterministic predictions of a future GLOF. We appreciate the reviewer’s comment, which has helped us improve the transparency of the uncertainty sources, their roles within the modelling framework, and their implications for interpretation of the results.
RC1-4 Response
We sincerely thank the reviewer for this valuable suggestion. We agree that a broader literature context is important for establishing the scientific relevance of the study beyond the Chubda Tsho study area.
- The literature review and discussion have been expanded to include additional studies on GLOF occurrence, hazard assessment, and hydrodynamic modelling in Bhutan and the wider Himalayan region. In particular, the revised discussion places the findings of the present study in relation to previous GLOF investigations from Bhutan and Nepal, with emphasis on flood-wave propagation, breach magnitude, downstream hydraulic response, flood arrival time, and the influence of steep and confined mountain-valley geometry.
- The revised discussion compares the simulated flood response with previous GLOF studies from Bhutan and Nepal, particularly in terms of peak discharge, flood-wave propagation, arrival time, inundation behaviour, and the influence of valley confinement. The comparison highlights both similarities and differences between the present results and previous studies and identifies the implications of these findings for understanding GLOF dynamics in confined Himalayan valleys. This comparison helps demonstrate that the study contributes not only a site-specific assessment of Chubda Tsho, but also a scenario-based evaluation of how breach magnitude and valley geometry influence downstream GLOF dynamics in a data-scarce Himalayan setting.
- We thank the reviewer for this recommendation, which has helped us strengthen the literature context, expand the discussion of previous GLOF studies from Bhutan and Nepal, and more clearly articulate the broader scientific contribution of the study.
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript. The comments have helped us clarify the scientific questions and contribution of the study, strengthen the description of the modelling framework and data sources, more explicitly address uncertainties and modelling limitations, and expand the discussion and comparison with previous GLOF studies in Bhutan and the wider Himalayan region. We believe that these revisions have substantially improved the clarity, transparency, scientific interpretation, and overall quality of the manuscript. We are grateful to the reviewer for the valuable suggestions and constructive guidance, which have helped us strengthen the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-2584-AC1
-
RC2: 'Comment on egusphere-2026-2584', Anonymous Referee #2, 24 Jun 2026
Summary
The study by Namgay et al. (2026) models three different glacial lake outburst flood (GLOF) scenarios originating from Chubda Tsho, located in the headwaters of the Chamkar Chu basin, using HEC-HMS and HEC-RAS. While the topic is highly relevant to local communities and disaster risk management in Bhutan, the manuscript currently offers limited scientific novelty in terms of methodology, analysis, and contribution to the existing body of knowledge on GLOF hazard and risk assessments. I encourage the authors to better articulate the novelty and significance of their work and to address the following comments before the manuscript can be considered for publication.
Major Comments
1. Chubda Tsho glacial lake characterization
The manuscript provides insufficient information on Chubda Tsho itself. The authors should include details on:
- The origin and formation history of the lake.
- Lake type (e.g., moraine-dammed, ice-contact, etc.).
- Historical lake expansion and development rates.
- The basis on which the lake was identified as a potentially dangerous glacial lake.
- Current lake area and volume, including methods used to derive these estimates.
- The area–volume relationship applied in the study.
- The reference year(s) used for lake area and volume estimates.
- Potential future lake expansion and corresponding volume increases.
- The likely maximum extent of future lake growth.
- In addition, more information on the parent glacier (Chubda Glacier) should be provided, including available estimates of glacier mass balance, velocity, thickness, and recent changes derived from remote sensing studies.
2. Flood and rainfall events used for calibration and validation
The manuscript provides insufficient information regarding the flood events used in the modelling framework (e.g., the 2007 flood, the 2015 event, and Cyclone Aila). The authors should clearly describe:
- The characteristics of each event.
- The hydrometeorological parameters used.
- Data sources and the agencies or organizations responsible for data collection.
- The rationale for selecting certain events for calibration and others for validation.
A more transparent explanation of the calibration and validation strategy is necessary.
3. Hydrograph assumptions and model validation
A fundamental difference exists between rainfall-induced flood hydrographs and GLOF hydrographs. The manuscript states that: “Hydrodynamic validation relied on the extreme rainfall-driven flood associated with Cyclone Aila (2009), in the absence of documented historical GLOF observations for the study basin.” This approach requires stronger justification because rainfall floods and moraine-dam breach floods are governed by different hydrological and hydraulic processes and often exhibit substantially different hydrograph shapes. I recommend validating the modelling framework against the well-documented 1994 Lugge Tsho GLOF in Bhutan, for which inundation extent and other observational datasets are available. Such validation would provide greater confidence in the model's ability to simulate GLOF dynamics.
Furthermore, the manuscript assumes that: “Synthetic outburst hydrographs were constructed assuming a single-peak breach process, with peak discharge occurring near the midpoint of breach development.” The authors should provide stronger justification for this assumption. Moraine-dam breach hydrographs are often characterized by a very steep rising limb and rapid peak discharge development. The current assumptions may not adequately capture the hydrograph characteristics of a moraine-dam failure. A comparison with documented GLOF hydrographs from similar moraine-dammed lakes would strengthen the analysis.
4. Downstream risk assessment
The study focuses primarily on flood propagation but provides limited assessment of downstream impacts. I recommend conducting a more comprehensive downstream risk assessment, similar to the approach adopted by Rinzin et al. (2023), including:
- Exposure of settlements and infrastructure.
- Potential impacts on hydropower facilities, roads, bridges, and agricultural land.
- Population at risk.
- Hazard and risk zonation.
Such an analysis would significantly enhance the practical relevance of the study.
5. Present-day and future GLOF scenarios
The manuscript would be substantially strengthened by incorporating future GLOF scenarios in addition to the present-day assessment. Given the ongoing expansion of Chubda Tsho, the authors should evaluate how future increases in lake area and volume may influence potential outburst magnitude, peak discharge, inundation extent, and downstream impacts. This could be achieved by projecting plausible future lake extents and volumes based on observed lake growth rates, glacier retreat, and topographic constraints.
Importantly, such an analysis could provide a clear element of scientific novelty for the study. While many existing GLOF assessments focus solely on present-day conditions, a comparative evaluation of present and future GLOF hazards would offer valuable insights into how climate-driven lake evolution may alter downstream flood risk over time. This would significantly enhance the study's relevance for long-term disaster risk reduction, infrastructure planning, and climate adaptation in the Chamkar Chu basin.
By explicitly comparing present-day and future hazard and risk scenarios, the authors would move beyond a conventional GLOF modelling exercise and provide a more forward-looking assessment of evolving cryospheric hazards.
6. Figure quality and visualization
All figures are currently below publication quality and require substantial improvement. In particular, Figures 3–7 and Figures 10–12 should be revised to improve:
- Resolution and readability.
- Map layout and cartographic quality.
- Color schemes and contrast.
- Label clarity and font size.
- Figure captions and explanatory content.
High-quality visualizations are essential for effectively communicating the modelling results and their implications.
Overall Recommendation
The study addresses an important hazard issue for the Chamkar Chu basin and downstream communities. However, the manuscript would be substantially strengthened by providing a more comprehensive characterization of Chubda Tsho and its parent glacier, improving the justification and validation of the modelling framework, incorporating future lake development and GLOF scenarios, expanding the downstream risk assessment, and enhancing figure quality. Most importantly, the authors should clearly demonstrate the scientific novelty and added value of the study relative to existing GLOF modelling research.
Rinzin, S., Zhang, G., Sattar, A., Wangchuk, S., Allen, S. K., Dunning, S., and Peng, M.: GLOF hazard, exposure, vulnerability, and risk assessment of potentially dangerous glacial lakes in the Bhutan Himalaya, J. Hydrol., 619, 129311, https://doi.org/10.1016/j.jhydrol.2023.129311, 2023.
Citation: https://doi.org/10.5194/egusphere-2026-2584-RC2 -
AC2: 'Reply on RC2', Tenzin Namgay, 27 Aug 2026
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript and for the valuable suggestions provided to strengthen the study. We appreciate the reviewer’s comments regarding the characterization of Chubda Tsho and its parent glacier, the hydrological and hydrodynamic modelling framework, validation strategy, uncertainty and hydrograph assumptions, downstream implications, future lake evolution, and figure quality. We have carefully considered all of the comments and suggestions and have revised the manuscript accordingly. Detailed responses to each comment are provided below.
RC2-1 Response
We sincerely thank the reviewer for highlighting the need for a more comprehensive characterization of Chubda Tsho and its parent glacier. We have revised the manuscript to provide clearer information on the lake characteristics, historical development, data sources, and limitations of the available observations.
- Chubda Tsho (Cham_gl 383 in the NCHM inventory) is located in the headwaters of the Chamkhar Chhu basin at approximately 4,868 m a.s.l. The lake is associated with an ice-cored moraine and a relatively flat threshold at the end moraine and is directly fed by Cham_gr 71. Earlier investigations reported dead ice beneath the moraine, active slides, and fresh ice cliffs around the lake and recommended continued monitoring and detailed investigation. These observations indicate a moraine-dammed glacial lake with evidence of ongoing glacier–moraine interaction and historical lake development. Based on these geomorphological and dam-related characteristics, Chubda Tsho was identified and retained as a potentially dangerous glacial lake (PDGL) in the NCHM assessment.
- Regarding historical lake development, NCHM (2019) reports a surface area of approximately 1.035 km² in 2001 and 1.388 km² in 2016, with the latter derived from Sentinel-2 imagery. The report also cites an earlier expansion rate of approximately 0.027 km² yr⁻¹ and indicates an approximately 564 m upward expansion between 2001 and 2016. These values represent historical observations and are not treated as direct projections of future lake growth.
- The lake area, storage, and maximum depth used in this study were obtained from NCHM (2019). The reported lake area is approximately 1.39 km² and corresponds to the 2016 NCHM inventory derived from Sentinel-2 imagery. The reported storage volume of 21.69 × 10⁶ m³ and maximum depth of 56 m were obtained from the bathymetric survey conducted in 2010 under the DGM–JICA/JST GLOF Project (2009–2012). No separate area–volume relationship was applied, as the storage estimate was already available from the bathymetric assessment. Differences in observation periods and the absence of repeated bathymetric surveys introduce uncertainty in the representation of present-day lake storage and geometry.
- The corrected DEM was also used to characterize the lake and outlet elevations, while the NCHM-reported lake storage and maximum depth were retained as the basis for defining the GLOF scenarios.
- Regarding future lake expansion, although historical expansion has been documented, the available observations do not provide a sufficiently constrained basis for projecting future lake area, volume, or maximum future extent. In particular, the available data do not provide sufficient information to establish a physically defensible future lake-growth trajectory or a reliable limiting lake extent. We therefore do not introduce an unsupported future-growth projection into the present simulations.
- For the parent glacier, NCHM identifies Cham_gr 71 as the glacier directly feeding Chubda Tsho. However, detailed site-specific observations of glacier mass balance, ice thickness, surface velocity, and recent glacier change were not available in the datasets used in this study. These parameters were therefore not incorporated into the hydrodynamic modelling and are acknowledged as data limitations and priorities for future investigation.
- Because site-specific breach observations and detailed geotechnical information for the moraine were unavailable, the potential outburst magnitude was represented using 50%, 75%, and 100% lake-storage release scenarios, corresponding to 10.845 × 10⁶, 16.2675 × 10⁶, and 21.69 × 10⁶ m³, respectively. Breach width, formation time, and peak discharge were estimated using the Froehlich (1995, 2008) empirical relationships, and the resulting hydrographs were constructed to conserve the specified outburst volumes.
We have revised the manuscript to clearly distinguish between historical observations, NCHM-derived lake parameters, modelling assumptions, and limitations associated with future lake evolution and glacier dynamics. We appreciate the reviewer’s comment, which has helped us improve the transparency and completeness of the Chubda Tsho characterization.
RC2-2 Response
We sincerely thank the reviewer for this comment. We agree that the roles of the different flood events and the calibration–validation strategy were not sufficiently clear in the original manuscript. We have revised the manuscript to clarify the characteristics, data sources, and distinct purposes of the 2007, 2015, and Cyclone Aila (2009) events.
- For the HEC-HMS hydrological model, observed rainfall and discharge data obtained from the National Center for Hydrology and Meteorology (NCHM), Bhutan, were used. The July 2007 flood event was selected for model calibration, while the August 2015 flood event was used for independent validation. These events were selected based on the availability of corresponding observed rainfall and discharge records, allowing the rainfall–runoff model to be calibrated and subsequently evaluated using an independent event. The discharge observations were obtained from the Chamkhar Chhu gauging station (Chamkhar; 90.754653° E, 27.549856° N), located approximately 47 km downstream of Chubda Tsho. The HEC-HMS model achieved NSE values of 0.90 for the 2007 calibration event and 0.56 for the 2015 validation event.
- The Cyclone Aila (2009) event was treated separately from the HEC-HMS calibration and validation events. It was selected specifically to evaluate the downstream HEC-RAS hydraulic setup, because an observed inundation extent for the Cyclone Aila flood was available in the downstream Chamkhar area from NCHM. The calibrated HEC-HMS model was used to generate the rainfall–runoff hydrograph for Aila, which was then applied as the upstream boundary condition of the downstream HEC-RAS model. The simulated inundation was compared with the observed Aila inundation extent, resulting in a CSI of 0.67 and an inundation-area difference of 2.56%. This evaluation was used to assess the downstream hydraulic representation and adopted Manning’s roughness coefficients.
- Importantly, the Cyclone Aila simulation was not used to calibrate or validate the Chubda Tsho GLOF simulations. The two applications have different purposes and upstream boundary conditions. For Cyclone Aila, the rainfall-driven hydrograph was applied to the downstream Chamkhar modelling domain. For the GLOF simulations, the upstream boundary condition was changed to the Chubda Tsho lake outlet, and synthetic breach hydrographs representing 50%, 75%, and 100% lake-storage release scenarios were applied.
- Because no documented historical GLOF has occurred at Chubda Tsho and corresponding observed GLOF discharge and inundation data are unavailable, direct calibration or validation of the Chubda Tsho GLOF simulations is not possible. The three GLOF scenarios were therefore designed as breach-sensitivity scenarios to systematically examine the downstream hydraulic response to different assumed release magnitudes.
The revised manuscript now provides greater detail on the event periods, observed rainfall and discharge data, hydrological and hydrodynamic model parameters, boundary conditions, and data sources, while clearly distinguishing the 2007 HEC-HMS calibration, 2015 HEC-HMS validation, 2009 Cyclone Aila downstream HEC-RAS evaluation, and separate Chubda Tsho GLOF scenario simulations. We appreciate the reviewer’s comment, which has helped us improve the transparency of the modelling and validation strategy.
RC2-3 Response:
We sincerely thank the reviewer for this important comment. We agree that rainfall-induced floods and moraine-dam breach floods are governed by different processes and can have substantially different hydrograph characteristics. We have therefore clarified the purpose of the Cyclone Aila assessment and the assumptions associated with the synthetic GLOF hydrographs.
- First, Cyclone Aila (2009) was not used to validate the GLOF breach process itself. It was used to independently evaluate the downstream HEC-RAS hydraulic setup under an observed extreme-flood condition, because an observed inundation extent was available for the downstream Chamkhar area. The Aila rainfall–runoff hydrograph generated using the calibrated HEC-HMS model was applied as the upstream boundary condition of the downstream HEC-RAS model, and the simulated inundation was compared with the observed Aila inundation extent. The simulation achieved a CSI of 0.67 and an inundation-area difference of 2.56%. This assessment was therefore intended to evaluate the representation of downstream hydraulic propagation, terrain, channel connectivity, and roughness, rather than to demonstrate equivalence between rainfall-flood and GLOF hydrographs.
- We appreciate the reviewer’s suggestion to use the 1994 Lugge Tsho GLOF as an additional reference. Although the 1994 Lugge Tsho event provides a valuable regional reference, Lugge Tsho is located in Punakha District, whereas Chubda Tsho is located in the Chamkhar Chhu basin of Bumthang District, and the two events occurred in different lake–river systems. Direct use of the Lugge Tsho event for validation of the present Chubda Tsho model would therefore require transferring observations between systems with differences in valley and river-channel geometry, terrain, LULC and associated Manning’s roughness, river perimeter, and model-domain configuration. The hydraulic model in this study was specifically developed for the Chubda Tsho–Chamkhar Chhu system. For this reason, we did not use the Lugge Tsho event as a direct validation case. Instead, the simulated GLOF responses were compared with reported results from documented Himalayan GLOF studies, including studies from Bhutan, to assess whether the magnitude and behaviour of the simulated responses are within physically plausible ranges. This comparison is treated as a plausibility assessment rather than formal validation.
- Regarding the synthetic GLOF hydrographs, no observed breach hydrograph or site-specific breach observations are available for Chubda Tsho. The breach characteristics were therefore estimated using the Froehlich (1995, 2008) empirical relationships from the available lake-storage and breach-depth information. For the 50%, 75%, and 100% release scenarios, the estimated breach formation times were approximately 0.68, 0.58, and 0.52 h, with corresponding estimated peak discharges of approximately 4,498, 8,382, and 13,035 m³ s⁻¹.
- The resulting hydrographs were constructed using an idealised two-part, single-peak formulation, consisting of a linear rising limb followed by an exponential recession limb. The peak was placed near the midpoint of the estimated breach-formation period as a simplified representation of the timing of peak discharge during breach development in the absence of observations defining the actual temporal evolution of the breach. The recession constant was then adjusted iteratively so that the integrated hydrograph volume matched the prescribed released lake volume for each scenario. The resulting final peak discharges were 4,412, 8,189, and 12,782 m³ s⁻¹ for the 50%, 75%, and 100% scenarios, respectively.
- We acknowledge that this idealised formulation cannot reproduce all characteristics of an actual moraine-dam failure, including potentially very rapid peak development, progressive breach enlargement, sediment entrainment, and erosion–deposition feedbacks. These processes could substantially influence the shape and magnitude of a real GLOF hydrograph. However, the necessary site-specific breach, sediment, and erosion observations are unavailable for Chubda Tsho. The synthetic hydrographs are therefore used consistently as scenario-based forcing conditions to examine the sensitivity of downstream hydraulic response to different assumed release magnitudes, rather than as deterministic predictions of an actual future breach hydrograph.
We have revised the manuscript to explicitly distinguish (i) the Cyclone Aila assessment of the downstream hydraulic model, (ii) the hypothetical Chubda Tsho breach-sensitivity scenarios, and (iii) comparison with published GLOF results as a plausibility check. The assumptions and limitations associated with the synthetic hydrograph formulation have also been clarified in the revised manuscript. We sincerely appreciate the reviewer’s comment, which has helped us improve the transparency and scientific interpretation of the modelling framework.
RC2-4 Response:
We sincerely thank the reviewer for this valuable suggestion. We agree that linking the simulated flood dynamics with downstream exposure is important for demonstrating the practical significance of the modelling results.
- The primary focus of the present manuscript is the hydrodynamic behaviour and breach-magnitude sensitivity of GLOF propagation from Chubda Tsho, including peak discharge, flood depth, flow velocity, arrival time, and inundation characteristics. A comprehensive downstream exposure, economic-loss, vulnerability, and evacuation assessment involves additional socioeconomic and asset-specific analyses and is therefore treated as a complementary component of the broader research.
- Nevertheless, in response to the reviewer’s suggestion, we have strengthened the revised manuscript by explicitly discussing the potential exposure of downstream settlements and critical infrastructure within the modelled floodplain, including roads, bridges, agricultural areas, and other major infrastructure. The spatial implications of the simulated inundation for the downstream Chamkhar area are also discussed in relation to the simulated hazard patterns.
- A quantitative population-at-risk assessment was not undertaken in the present manuscript because sufficiently detailed spatially disaggregated population data were not available. Similarly, detailed economic-loss and vulnerability estimates require asset-specific valuation and vulnerability information that could not be reliably constrained using the available datasets. We have therefore avoided presenting potentially uncertain estimates as definitive risk values.
- The revised manuscript now more clearly distinguishes between the hydrodynamic hazard assessment presented here and complementary downstream exposure and risk analyses. This allows the present study to maintain its primary focus on hydrodynamic behaviour and breach-magnitude sensitivity while providing appropriate discussion of the practical implications of the simulated GLOF hazard.
The revised manuscript now more clearly distinguishes between the hydrodynamic hazard assessment presented here and the complementary downstream exposure and risk analyses. We appreciate the reviewer’s recommendation, which has helped us strengthen the discussion of the practical implications of the simulated GLOF hazard while maintaining the scientific focus of the manuscript.
RC2-5 Response:
We sincerely thank the reviewer for this valuable suggestion. We agree that future lake evolution is an important consideration for long-term GLOF hazard assessment, particularly given the documented historical expansion of Chubda Tsho.
- As clarified in the revised manuscript, historical lake expansion has been documented, with the lake area increasing from approximately 1.035 km² in 2001 to 1.388 km² in 2016. However, the available observations do not provide a sufficiently constrained basis for reliably projecting future lake area, storage volume, or maximum lake extent. In particular, consistent multi-temporal bathymetric information and site-specific data on glacier mass balance, ice thickness, and glacier dynamics are not available.
- We therefore considered it inappropriate to extrapolate the historical expansion rate directly to construct future lake scenarios, as this would introduce additional unsupported assumptions into the breach and hydrodynamic modelling. Instead, the present study focuses on scenario-based sensitivity to different outburst magnitudes using the currently available lake storage, represented by the 50%, 75%, and 100% release scenarios. These scenarios are intended to examine the sensitivity of downstream flood dynamics to increasing release magnitude and are not intended to represent future climate scenarios or probabilistic predictions.
- We have clarified this distinction in the revised manuscript and explicitly acknowledged future lake expansion and glacier evolution as important uncertainties and priorities for future research. A physically constrained assessment of future GLOF hazards would require additional multi-temporal lake and glacier observations, repeated bathymetric surveys, and coupled glacier–lake evolution modelling.
We sincerely appreciate the reviewer’s recommendation, which has helped us clarify the scope and limitations of the present study and identify future lake evolution as an important direction for extending GLOF hazard assessment in the Chamkhar Chhu basin.
RC2-6 Response:
We sincerely thank the reviewer for this helpful comment. We have carefully revised the figures throughout the manuscript to improve their resolution, readability, cartographic layout, colour contrast, label visibility, font sizes, and overall presentation quality. The map layouts have also been refined to improve the clarity of spatial information, including appropriate legends, scale bars, north arrows, and coordinate information where applicable. Figure captions have been revised to provide clearer and more informative descriptions of the information presented.
- The revised figures provide clearer visualization of the hydrodynamic model setup, GLOF scenarios, flood propagation, and downstream hydraulic results, with improved consistency in formatting and presentation throughout the manuscript.
- We sincerely appreciate the reviewer’s constructive suggestion, which has helped us improve the overall visual quality, clarity, and readability of the manuscript.
We sincerely thank the reviewer for the careful and constructive evaluation of our manuscript. The reviewer’s comments have helped us substantially improve the characterization of Chubda Tsho and its parent glacier, clarify the hydrological and hydrodynamic modelling and validation framework, strengthen the discussion of hydrograph assumptions and uncertainties, better articulate the downstream implications of the simulated GLOF hazard, clarify the scope and limitations of the present-day scenario analysis, and substantially improve the quality and presentation of the figures. We believe that these revisions have improved the clarity, transparency, scientific interpretation, and overall quality of the manuscript, while maintaining its primary focus on breach-magnitude sensitivity and downstream GLOF dynamics in the Chubda Tsho–Chamkhar Chhu system. We are sincerely grateful to the reviewer for the valuable recommendations and constructive guidance, which have helped us strengthen the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-2584-AC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 101 | 49 | 21 | 171 | 13 | 8 |
- HTML: 101
- PDF: 49
- XML: 21
- Total: 171
- BibTeX: 13
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review egusphere-2026-2584
The authors have chosen one of the lakes identified as potentially hazardous in Bhutan for a hydrodynamic simulation to assess flood extents under a variety of breach scenarios. They use the HEC modelling suite for both the estimated discharge peaks from a hydrological model (if I understand correctly) as well as the hydrodynamical modelling and come to the general conclusion that flood waves in the Chamkar Valley are rapid, leaving little time for early warning. In general, such studies are interesting and required to get a better understanding of potential risks and general preparedness. The manuscript is also very well written, and the figures are generally clear, hence leaving hardly anything for me to comment on in detail. However, there are four crucial shortcomings, that lead me to suggest a thorough rethinking of the scope before resubmitting.