the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Apr 2023
Multi-scale variations of hydro-mechanical conditions at the base of the surge-type glacier Kongsvegen, Svalbard
Abstract. Fast glacier flow and dynamic instabilities, such as surges, are primarily caused by changes at the ice-bed interface, where basal slip and sediment deformation drive basal glacier motion. Determining subglacial conditions and their responses to hydraulic forcing (e.g. rainfall, surface melt) remains challenging due to the difficulty of accessing the glacier bed. In this study, we monitor the interplay between surface runoff and hydro-mechanical conditions at the base of the Arctic surge-type glacier Kongsvegen, in Svalbard, over two contrasting melt seasons. Kongsvegen last surged in 1948, after which it entered a prolonged quiescent phase. Around 2014, flow speeds began to increase, sign of an imminent new fast-flow event. In 2021 we instrumented a borehole to assess subglacial conditions at the local scale and deployed seismometers to monitor the subglacial conditions at the kilometer scale. We measure both subglacial water pressure within the borehole with a water pressure sensor and till rheology with a ploughmeter inserted into the sediments at the bottom of the borehole. We use channel-flow-induced tremors recorded by a seismometer to characterize hydraulic conditions over a kilometre scale at the base of the glacier. The records cover the period from spring 2021 until summer 2022. To characterize the variations in the subglacial conditions caused by changes in surface runoff, we investigate the phase relationship (i.e. how two variables evolve in time) of the following hydro-mechanical condition proxies: water pressure, hydraulic gradient, hydraulic radius, and sediment ploughing forces. We analyse these proxies versus modelled runoff analyzed over seasonal, multi-day and diurnal time-scales. We compare our results with existing theories in terms of subglacial drainage system evolution and sediment shear strength to describe various aspects of subglacial conditions. We find apparent ambiguities in the interpretation of different variables recorded by individual sensors, thus demonstrating the importance of using multi-sensor records in a multi-scale analysis. This study highlights the different adaption of the subglacial drainage system during short, low melt intensity season in 2021, against long, high intensity melt season in 2022. In the short and low intensity melt season, we find that the subglacial drainage system evolves at equilibrium with runoff, increasing its capacity as the melt season progresses. In contrast, during the long and high intensity melt season 2022, we find that the subglacial drainage system evolves transiently to respond to the abrupt and high intensity input of precipitation and melt water conveyed to the bed. In this configuration, the subglacial channels evolution is not rapid enough to adapt immediately to the forcing conditions. The drainage capacity of the main active channels is exceeded, promoting the water to leak in poorly connected areas of the bed, increasing the water pressure, resulting in speed-up events. Another robust outcome of our analysis is, that, on a seasonal scale, till shear strength variations are mainly anti-correlated with water pressure variations (consistent with a Coulomb-plastic behavior), whereas on shorter time scales especially during speed-up events, the two variables correlate, describing a viscous rheology. To our knowledge, such contrasted behaviors of the sediment rheology and subglacial flow at the base of a glacier have not been reported before.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(4931 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4931 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-618', Anonymous Referee #1, 09 Jun 2023
In this paper the authors explore the relationships between subglacial water pressure, seismic power, ploughmeter force, surface velocity and modelled surface meltwater input for a large surge-type glacier in Svalbard for the 2021 and 2022 melt seasons. Based on recently established relationships between seismic power and discharge by Gimbert et al (2016), the authors use the seismic power data and modelled meltwater input to derive channel cross section and pressure gradient to understand the evolution of these variables for the two contrasting melt seasons. The derived and measured subglacial variables are filtered into diurnal, multi-day and seasonal windows and plotted against each other in phase space with implicit dependence on time. From the phase space relations and the filtered signals, the authors categorize the time periods into four domains to understand the state of channelized flow in relation to Rothlisberger theory for steady channel flow. They have identified periods of time when channel flow is in steady state vs transient. Unrelated to the seismic power, they explain ploughmeter and water pressure data by proposing that sometimes till is coulomb-plastic and sometimes it is viscous. The authors have carried out an impressive campaign to instrument and analyze a rich dataset and their work will contribute to a better understanding of subglacial processes.
The major concerns that I have with the work are listed here while minor comments are annotated on the attached PDF and listed below.
For much of the paper, the references are not accurate or pertinent. I stopped commenting on the references halfway through the introduction, but the authors should consider revising the appropriateness of references throughout the paper. In general, the authors need to consider using more original citations and be sure that the citation supports the statement.
For a discussion of subglacial processes, the authors need to do substantially more work to fit their observations with the literature, especially for till mechanics. My expertise is not in glacier seismicity so I cannot offer much feedback that way. My personal take on till mechanics is that, as a community, we’ve moved passed the idea of a viscous rheology for till. There are many basal processes related to till mechanics that are not considered and could really change the interpretation of the results, including ice-till coupling, cavitation around clasts within the till, sheet flow at the ice-till interface, regelation infiltration, water pressure fluctuations in the till, etc. The work focuses heavily on subglacial hydrology through the lens of channel flow and channel geometry, and in the discussion the authors recognize that the water pressure and ploughmeter data cannot be explained by channel flow characteristics and that distributed flow is probably important. It would therefore make sense that the authors do more to fit all results, including seismic power, into existing theories that describe the seasonal evolution of the drainage system and the relationships between channelized and distributed flow. At present, all observations are being analyzed through the lens of channel flow where the number of channels (N) is fixed. The authors should focus more on explaining the measured variables rather than the derived quantities.
As mentioned above, the paper relies heavily on the interpretation of derived variables (channel cross section and pressure gradient) and discusses the variables as though they’re measured quantities. It is intriguing that the phase space portraits can show log-linear correlations that correspond to end-member states of fixed geometry or fixed gradient, but the authors should approach this with caution. There may be many reasons for log-linear correlations amongst variables and the limitations to the equations used are not discussed with relevant detail.
- Bedload transport and fluvial erosion should play a very important role in generating noise, and bedload transport can show its own hysteresis with discharge.
- The Q being plotted against subglacial variables is also derived and does not represent the discharge for the Rothlisberger equation. The time delay and changes in the time delay throughout the season should have a very large impact on the diurnal filter window.
- One of my main concerns is that all the equations are describing the channel evolution at a point. At line 495 the authors state that noise is picked up from a 1km^2 area and detecting the loudest noise, but wouldn’t the noise be integrated over this area where channels can show a large variation is size and shape, through time and space?
- Using the seasonal averaging window, the filtered data suggests that channels are in a transient state for a long period of time, so why would daily or multi-day time windows be used over the same time period investigate the possibility of steady state?
- How do the authors address the noise generated from open channel flow vs. pipe full flow? (perhaps this is addressed and I misunderstood)
- The x axis for derived quantities is labelled as log (X/Xref) AND the scale is log, does that make sense? The x axis should show units.
The authors have an interesting dataset pertaining to hydraulic connectivity and basal sliding where continuous till is inferred. The paper would be a lot stronger if there was less focus on resolving the characteristics of channel flow that are derived with seismic noise, while focusing more on explaining the relationships amongst measured quantities and explaining the results based on a stronger foundation in existing literature for basal processes.
Another minor comment is that the intro to the paper focuses a lot on surge-type glaciers and the basal processes of surge-type glaciers, but a discussion on the mechanics of surge-type glaciers is absent. Do the observations inform us of anything new about the dynamics of surging glaciers and how do the results fit in the context of this glacier building up to a full surge?
- AC1: 'Reply on RC1', Coline BOUCHAYER, 25 Nov 2023
- AC2: 'Reply on RC1', Coline BOUCHAYER, 25 Nov 2023
-
RC2: 'Comment on egusphere-2023-618', Anonymous Referee #2, 08 Aug 2023
Review for Bouchayer et al., ‘Multi-scale variations of hydro-mechanical conditions at the base of the surge-type glaciers Kongsvegen, Svalbard’
The manuscript by Bouchayer et al. describes a series of novel subglacial and glacier surface observations collected on Kongsvegen Glacier during 2021 and 2022. These observations, including subglacial pressure, ploughing force, seismic power, ice surface velocity, and surface meltwater runoff are used to characterize the behavior of the subglacial system during two contrasting melt years. To do this, the authors utilize seismic processing methods to derive subglacial channel characteristics and classify the relationship between meltwater runoff and variables descriptive of the subglacial environment over seasonal, event, and daily timescales. Overall, they conclude that during the low melt year (2021), the subglacial system was able to readily adapt to changes in runoff availability, while during the high melt year (2022), runoff variability frequently overwhelmed the efficient subglacial system resulting in ice velocity acceleration events. However, the relationships examined on shorter timescales and those related to till behavior suggest a complicated and time evolving subglacial environment.
The observations presented within Bouchayer et al. are novel and the derived relationships between a range of variables on multiple timescales are thought provoking. In addition, the quantitative approach to characterizing the relationship between different variables is commendable. The manuscript is generally well written and conveys the complexity associated with multiple observations and over multiple time and space scales. However, I do wonder if the ambiguity in the results is more related to how the methods were applied than in the observations themselves. As such, I have some concerns about how certain methods were applied and some suggestions regarding analysis and interpretation of the datasets. Below are general comments pertaining to analysis and interpretation, comments about the manuscript structure, and line and figure comments.
General comments on analysis and interpretation
- Much of what is being described (as illustrated in Figure 10), is reminiscent of the ‘preferential drainage axes’ of Haut Glaicer d’Arolla (e.g., Sharp et al. 1993; Mair et al. 2001; Mair et al., 2003), and it would be highly relevant to include a discussion of PDAs when considering how the Kongsvegen surface velocities, relate to subglacial pressure, seismic power, and till behavior.
- Overall, there is very little discussion of surface ice velocities. Ultimately understanding the subglacial system is necessary to inform our understanding of glacier motion, surges, seasonal, etc. More discussion of the link between subglacial conditions and surface velocities is warranted, including plots examining the relationship F, Q, & p vs us. An exploration of F, us, and uplift really is warrented.
- The seismic processing methods require a number of assumptions that may or may not be met by the data. Indeed, the authors state as much on line 472. Kongsvegen is not hard-bedded and the assumptions that the number of channels is constant and that these channels are full are not necessarily met suggest that the calculation of the hydraulic gradient and radius are not robust. This issue in addition to the one below suggests that the authors should consider a simpler way to consider seismic power and its relation to both the surface forcing and ice motion.
- The authors are correct in using ‘Runoff’ for modeled glacier surface runoff, but the abbreviation Q is somewhat misleading as Q is typically shorthand for Discharge – which is the volume of water that passes through a cross section per unit time. However, my main concern on this point is that the modeled surface runoff is used to calculate the hydraulic radius and hydraulic gradient within inferred subglacial channels. Gimbert et al. (2016) and Nanni et al. (2020) calculate the discharge of the subglacial channels using the Manning Strickler equation. To argue that surface runoff = subglacial channel discharge, a number of assumptions are made, including that all surface runoff flows within subglacial channels (or at least flows turbulently), for the entire time period, within a region that can be monitored by the seismic station. It is quite possible that this assumed equivalency between surface runoff and subglacial channel discharge can at least partly explain the ambiguity of the results, and the authors should carefully consider whether the calculations of R and S are robust enough to use in the analysis.
- I am somewhat surprised that the borehole is in an active part of the drainage system during relatively low melt year, but in an inactive part of the drainage system during the high melt year – would typically be reversed and patterns of channelization tend to be consistent, though more or less extensive, from year to year. Could this be due to the initial hot water drilling? Changes in till characteristics? It would be nice to see the explanation. It is hard to see where the borehole would be located, both theoretically and in Figure 10, based on the analysis.
- The combined use of reanalysis and forecast data give me pause. While both use a similar model configuration and give regionally similar results, there are differences in the forcings and configurations that could impact temperature (including SW and LW radiation) and precipitation and there are documented local differences between the two systems. Kongsvegen has a weather station; could confirmation of similarity or differences be determined? Alternatively, because CARRA ends in 2021, could the AROME-Arctic analysis (not forecasts, I believe the analysis is the MET Nordic Analysis) be used for both years. Whatever tack is taken, more clarity on any differences between the met forcings used for CryoGrid between each year need to be included.
General comments on manuscript structure
- The title, abstract and introduction emphasize that Kongsvegen is a surge type glacier, but there is no discussion of how the observations inform our understanding of surge mechanics. Either the thrust of the Discussion needs to change, or the introductory materials should be adjusted.
- I empathize with the authors’ desire to be succinct, but using many abbreviations and numbers to identify different classes makes reading the discussion challenging and requires multiple references to previous figures and text. Could more descriptive terms for the different classes be used?
- I’d like to see the number and breadth of references expanded. There are multiple areas where there are no pertinent references.
Line comments
L10. I would add here that this information is used to derive hydraulic gradient, and subglacial channel hydraulic radius.
L13. Consider being specific here: water pressure and force and measured, hydraulic gradient and hydraulic radius are inferred/calculated from previously determined relationships.
L14. modeled surface runoff
- The ambiguity seems to be mostly between the inferred variables vs the measured variables (e.g. direct vs seismically inferred). It might be worth being more specific here.
L35. Add an ‘e.g.’ to this citation. There are many papers suggesting this.
L38. Add an ‘e.g.’ to this citation.
L39-48. Surges are an interesting transient event that can be used to understand basal conditions, but they are not discussed within the content of the observations. Consider revising this paragraph to be more general.
L60ish. It’s worth mentioning isolated cavities since these are invoked later in the manuscript (e.g., Iken et al., 1983).
L84. ‘Truffer’
L92. New paragraph.
- Is the geophone installed in the borehole? It’s unclear. It might be worth including a subheading ‘near surface instrumentation.’
L154-160. The position information at least need general uncertainties.
L162. Westermann et al. (2023)?
L165. See my general note. Also, is the forecast the ensemble mean or the single unperturbed member?
L170. This is a big assumption. Is there any justification for this that could be included?
L197. In theory, I don’t think there is a problem assuming a constant number of channels for short periods of time, but one of the final conclusions is that the borehole is connected to the efficient system in 2021 and in an isolated(ish) region in 2022, suggesting a different number of channels.
L244. velocity should have the abbreviation us.
L254. Only total precipitation is included on Figure 4. It would be useful to have both rain and snow fall.
L257. Sometimes the second number in the figure references is circled and sometimes it isn’t.
L293. If dates are referenced here, they should be clearly identifiable in Figure 5.
L304. Figure 5i doesn’t seem to indicate a linear relationship… perhaps this is because the axis ranges are vastly different.
L306. The figure seems misplaced.
L344. What is the overburden pressure at the borehole location? The lack of diurnal variability in p and F and us, suggests that any subglacial channels are not completely water filled except during melt/rain events. This has a number of implications for the analysis.
L365. A constant R would be expected, if the channel is water filled. The lack of diurnal variations suggests that this might not be true. In a partially filled channel, R would increase with increasing S.
L363-381. Some references would be beneficial.
L429. p didn’t exhibit diurnal variations, so this statement seems a bit misleading.
L430. Could this rapid adjustment of R be the result of subglacial channels that are not filled?
L460. There seem to be more diurnal variations in p during 2022 than in 2021, indeed it looks like at least 60% of the days have enough variability to assign a class.
L472 See general note.
L497. See the literature on preferential drainage axis. This is what is being described in Figure 10.
L555. One thing to consider is how the behavior illustrated in Figure 10 transfers mechanical support of the overlying ice and how that might impact till behavior or measured force on the plough meter.
Figures
Figure 1
- Data source for panel b?
Figure 2 could easily be in an appendix.
Figure 3
- The class colors here and in the other figures are hard to distinguish, could the be more distinct?
- It would be useful to have the same color scale as in Figure 5, etc.
Figure 4.
- Rainfall is discussed multiple times in the text, so rainfall and snowfall should be parsed in panel a.
- The winter period isn’t analyzed, could it be cut out (and possibly included in the Appendices) to make the summer seasons bigger?
- I don’t see any blue or grey shaded areas on my printed version.
- Are there diurnal variations in ice velocity?
Figure 5
- It would be useful to include how to read figure 5 a-c, f-h in the caption including how the curves relate to the bounds to determine behavior.
- Scale the color bars to be the same number of days such that it’s clear that the 2022 data doesn’t go to the end of the melt season and they should be the same across Figures (right now Figure 7 has a different color scale for 2022).
- The vastly different ranges on the x and y axes make it difficult to interpret behavior (see line comment 304). These should be standardized as much as possible, in ways that highlight the main points of the analysis.
Figure 6
- Could the windows be plotted on subpanels b and d?
Figure 7
- See notes about color scale and axes for Figure 5.
Figure 9
How are the melt seasons combined in panel a?
Figure 10
- Where would the borehole sit in the subglacial plan view maps?
References (not already included in the manuscript)
Mair, D., Nienow, P., Willis, I. and Sharp, M.: Spatial patterns of glacier motion during a high-velocity event: Haut Glacier d’Arolla, Switzerland, Journal of Glaciology, 47(156), 9–20, doi:10.3189/172756501781832412, 2001.
Mair, D., Willis, I., Fischer, U. H., Hubbard, B., Nienow, P. and Hubbard, A.: Hydrological controls on patterns of surface, internal and basal motion during three “‘spring events’”: Haut Glacier d’Arolla, Switzerland, Journal of Glaciology, 49(167), 555–567, doi:10.3189/172756503781830467, 2003.
Sharp, M., Richards, K., Willis, I., Arnold, N., Nienow, P., Lawson, W. and Tison, J.-L.: Geometry, bed topography and drainage system structure of the haut glacier d’Arolla, Switzerland, Earth Surf. Process. Landforms, 18(6), 557–571, doi:10.1002/esp.3290180608, 1993.
Citation: https://doi.org/10.5194/egusphere-2023-618-RC2 - AC3: 'Reply on RC2', Coline BOUCHAYER, 25 Nov 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-618', Anonymous Referee #1, 09 Jun 2023
In this paper the authors explore the relationships between subglacial water pressure, seismic power, ploughmeter force, surface velocity and modelled surface meltwater input for a large surge-type glacier in Svalbard for the 2021 and 2022 melt seasons. Based on recently established relationships between seismic power and discharge by Gimbert et al (2016), the authors use the seismic power data and modelled meltwater input to derive channel cross section and pressure gradient to understand the evolution of these variables for the two contrasting melt seasons. The derived and measured subglacial variables are filtered into diurnal, multi-day and seasonal windows and plotted against each other in phase space with implicit dependence on time. From the phase space relations and the filtered signals, the authors categorize the time periods into four domains to understand the state of channelized flow in relation to Rothlisberger theory for steady channel flow. They have identified periods of time when channel flow is in steady state vs transient. Unrelated to the seismic power, they explain ploughmeter and water pressure data by proposing that sometimes till is coulomb-plastic and sometimes it is viscous. The authors have carried out an impressive campaign to instrument and analyze a rich dataset and their work will contribute to a better understanding of subglacial processes.
The major concerns that I have with the work are listed here while minor comments are annotated on the attached PDF and listed below.
For much of the paper, the references are not accurate or pertinent. I stopped commenting on the references halfway through the introduction, but the authors should consider revising the appropriateness of references throughout the paper. In general, the authors need to consider using more original citations and be sure that the citation supports the statement.
For a discussion of subglacial processes, the authors need to do substantially more work to fit their observations with the literature, especially for till mechanics. My expertise is not in glacier seismicity so I cannot offer much feedback that way. My personal take on till mechanics is that, as a community, we’ve moved passed the idea of a viscous rheology for till. There are many basal processes related to till mechanics that are not considered and could really change the interpretation of the results, including ice-till coupling, cavitation around clasts within the till, sheet flow at the ice-till interface, regelation infiltration, water pressure fluctuations in the till, etc. The work focuses heavily on subglacial hydrology through the lens of channel flow and channel geometry, and in the discussion the authors recognize that the water pressure and ploughmeter data cannot be explained by channel flow characteristics and that distributed flow is probably important. It would therefore make sense that the authors do more to fit all results, including seismic power, into existing theories that describe the seasonal evolution of the drainage system and the relationships between channelized and distributed flow. At present, all observations are being analyzed through the lens of channel flow where the number of channels (N) is fixed. The authors should focus more on explaining the measured variables rather than the derived quantities.
As mentioned above, the paper relies heavily on the interpretation of derived variables (channel cross section and pressure gradient) and discusses the variables as though they’re measured quantities. It is intriguing that the phase space portraits can show log-linear correlations that correspond to end-member states of fixed geometry or fixed gradient, but the authors should approach this with caution. There may be many reasons for log-linear correlations amongst variables and the limitations to the equations used are not discussed with relevant detail.
- Bedload transport and fluvial erosion should play a very important role in generating noise, and bedload transport can show its own hysteresis with discharge.
- The Q being plotted against subglacial variables is also derived and does not represent the discharge for the Rothlisberger equation. The time delay and changes in the time delay throughout the season should have a very large impact on the diurnal filter window.
- One of my main concerns is that all the equations are describing the channel evolution at a point. At line 495 the authors state that noise is picked up from a 1km^2 area and detecting the loudest noise, but wouldn’t the noise be integrated over this area where channels can show a large variation is size and shape, through time and space?
- Using the seasonal averaging window, the filtered data suggests that channels are in a transient state for a long period of time, so why would daily or multi-day time windows be used over the same time period investigate the possibility of steady state?
- How do the authors address the noise generated from open channel flow vs. pipe full flow? (perhaps this is addressed and I misunderstood)
- The x axis for derived quantities is labelled as log (X/Xref) AND the scale is log, does that make sense? The x axis should show units.
The authors have an interesting dataset pertaining to hydraulic connectivity and basal sliding where continuous till is inferred. The paper would be a lot stronger if there was less focus on resolving the characteristics of channel flow that are derived with seismic noise, while focusing more on explaining the relationships amongst measured quantities and explaining the results based on a stronger foundation in existing literature for basal processes.
Another minor comment is that the intro to the paper focuses a lot on surge-type glaciers and the basal processes of surge-type glaciers, but a discussion on the mechanics of surge-type glaciers is absent. Do the observations inform us of anything new about the dynamics of surging glaciers and how do the results fit in the context of this glacier building up to a full surge?
- AC1: 'Reply on RC1', Coline BOUCHAYER, 25 Nov 2023
- AC2: 'Reply on RC1', Coline BOUCHAYER, 25 Nov 2023
-
RC2: 'Comment on egusphere-2023-618', Anonymous Referee #2, 08 Aug 2023
Review for Bouchayer et al., ‘Multi-scale variations of hydro-mechanical conditions at the base of the surge-type glaciers Kongsvegen, Svalbard’
The manuscript by Bouchayer et al. describes a series of novel subglacial and glacier surface observations collected on Kongsvegen Glacier during 2021 and 2022. These observations, including subglacial pressure, ploughing force, seismic power, ice surface velocity, and surface meltwater runoff are used to characterize the behavior of the subglacial system during two contrasting melt years. To do this, the authors utilize seismic processing methods to derive subglacial channel characteristics and classify the relationship between meltwater runoff and variables descriptive of the subglacial environment over seasonal, event, and daily timescales. Overall, they conclude that during the low melt year (2021), the subglacial system was able to readily adapt to changes in runoff availability, while during the high melt year (2022), runoff variability frequently overwhelmed the efficient subglacial system resulting in ice velocity acceleration events. However, the relationships examined on shorter timescales and those related to till behavior suggest a complicated and time evolving subglacial environment.
The observations presented within Bouchayer et al. are novel and the derived relationships between a range of variables on multiple timescales are thought provoking. In addition, the quantitative approach to characterizing the relationship between different variables is commendable. The manuscript is generally well written and conveys the complexity associated with multiple observations and over multiple time and space scales. However, I do wonder if the ambiguity in the results is more related to how the methods were applied than in the observations themselves. As such, I have some concerns about how certain methods were applied and some suggestions regarding analysis and interpretation of the datasets. Below are general comments pertaining to analysis and interpretation, comments about the manuscript structure, and line and figure comments.
General comments on analysis and interpretation
- Much of what is being described (as illustrated in Figure 10), is reminiscent of the ‘preferential drainage axes’ of Haut Glaicer d’Arolla (e.g., Sharp et al. 1993; Mair et al. 2001; Mair et al., 2003), and it would be highly relevant to include a discussion of PDAs when considering how the Kongsvegen surface velocities, relate to subglacial pressure, seismic power, and till behavior.
- Overall, there is very little discussion of surface ice velocities. Ultimately understanding the subglacial system is necessary to inform our understanding of glacier motion, surges, seasonal, etc. More discussion of the link between subglacial conditions and surface velocities is warranted, including plots examining the relationship F, Q, & p vs us. An exploration of F, us, and uplift really is warrented.
- The seismic processing methods require a number of assumptions that may or may not be met by the data. Indeed, the authors state as much on line 472. Kongsvegen is not hard-bedded and the assumptions that the number of channels is constant and that these channels are full are not necessarily met suggest that the calculation of the hydraulic gradient and radius are not robust. This issue in addition to the one below suggests that the authors should consider a simpler way to consider seismic power and its relation to both the surface forcing and ice motion.
- The authors are correct in using ‘Runoff’ for modeled glacier surface runoff, but the abbreviation Q is somewhat misleading as Q is typically shorthand for Discharge – which is the volume of water that passes through a cross section per unit time. However, my main concern on this point is that the modeled surface runoff is used to calculate the hydraulic radius and hydraulic gradient within inferred subglacial channels. Gimbert et al. (2016) and Nanni et al. (2020) calculate the discharge of the subglacial channels using the Manning Strickler equation. To argue that surface runoff = subglacial channel discharge, a number of assumptions are made, including that all surface runoff flows within subglacial channels (or at least flows turbulently), for the entire time period, within a region that can be monitored by the seismic station. It is quite possible that this assumed equivalency between surface runoff and subglacial channel discharge can at least partly explain the ambiguity of the results, and the authors should carefully consider whether the calculations of R and S are robust enough to use in the analysis.
- I am somewhat surprised that the borehole is in an active part of the drainage system during relatively low melt year, but in an inactive part of the drainage system during the high melt year – would typically be reversed and patterns of channelization tend to be consistent, though more or less extensive, from year to year. Could this be due to the initial hot water drilling? Changes in till characteristics? It would be nice to see the explanation. It is hard to see where the borehole would be located, both theoretically and in Figure 10, based on the analysis.
- The combined use of reanalysis and forecast data give me pause. While both use a similar model configuration and give regionally similar results, there are differences in the forcings and configurations that could impact temperature (including SW and LW radiation) and precipitation and there are documented local differences between the two systems. Kongsvegen has a weather station; could confirmation of similarity or differences be determined? Alternatively, because CARRA ends in 2021, could the AROME-Arctic analysis (not forecasts, I believe the analysis is the MET Nordic Analysis) be used for both years. Whatever tack is taken, more clarity on any differences between the met forcings used for CryoGrid between each year need to be included.
General comments on manuscript structure
- The title, abstract and introduction emphasize that Kongsvegen is a surge type glacier, but there is no discussion of how the observations inform our understanding of surge mechanics. Either the thrust of the Discussion needs to change, or the introductory materials should be adjusted.
- I empathize with the authors’ desire to be succinct, but using many abbreviations and numbers to identify different classes makes reading the discussion challenging and requires multiple references to previous figures and text. Could more descriptive terms for the different classes be used?
- I’d like to see the number and breadth of references expanded. There are multiple areas where there are no pertinent references.
Line comments
L10. I would add here that this information is used to derive hydraulic gradient, and subglacial channel hydraulic radius.
L13. Consider being specific here: water pressure and force and measured, hydraulic gradient and hydraulic radius are inferred/calculated from previously determined relationships.
L14. modeled surface runoff
- The ambiguity seems to be mostly between the inferred variables vs the measured variables (e.g. direct vs seismically inferred). It might be worth being more specific here.
L35. Add an ‘e.g.’ to this citation. There are many papers suggesting this.
L38. Add an ‘e.g.’ to this citation.
L39-48. Surges are an interesting transient event that can be used to understand basal conditions, but they are not discussed within the content of the observations. Consider revising this paragraph to be more general.
L60ish. It’s worth mentioning isolated cavities since these are invoked later in the manuscript (e.g., Iken et al., 1983).
L84. ‘Truffer’
L92. New paragraph.
- Is the geophone installed in the borehole? It’s unclear. It might be worth including a subheading ‘near surface instrumentation.’
L154-160. The position information at least need general uncertainties.
L162. Westermann et al. (2023)?
L165. See my general note. Also, is the forecast the ensemble mean or the single unperturbed member?
L170. This is a big assumption. Is there any justification for this that could be included?
L197. In theory, I don’t think there is a problem assuming a constant number of channels for short periods of time, but one of the final conclusions is that the borehole is connected to the efficient system in 2021 and in an isolated(ish) region in 2022, suggesting a different number of channels.
L244. velocity should have the abbreviation us.
L254. Only total precipitation is included on Figure 4. It would be useful to have both rain and snow fall.
L257. Sometimes the second number in the figure references is circled and sometimes it isn’t.
L293. If dates are referenced here, they should be clearly identifiable in Figure 5.
L304. Figure 5i doesn’t seem to indicate a linear relationship… perhaps this is because the axis ranges are vastly different.
L306. The figure seems misplaced.
L344. What is the overburden pressure at the borehole location? The lack of diurnal variability in p and F and us, suggests that any subglacial channels are not completely water filled except during melt/rain events. This has a number of implications for the analysis.
L365. A constant R would be expected, if the channel is water filled. The lack of diurnal variations suggests that this might not be true. In a partially filled channel, R would increase with increasing S.
L363-381. Some references would be beneficial.
L429. p didn’t exhibit diurnal variations, so this statement seems a bit misleading.
L430. Could this rapid adjustment of R be the result of subglacial channels that are not filled?
L460. There seem to be more diurnal variations in p during 2022 than in 2021, indeed it looks like at least 60% of the days have enough variability to assign a class.
L472 See general note.
L497. See the literature on preferential drainage axis. This is what is being described in Figure 10.
L555. One thing to consider is how the behavior illustrated in Figure 10 transfers mechanical support of the overlying ice and how that might impact till behavior or measured force on the plough meter.
Figures
Figure 1
- Data source for panel b?
Figure 2 could easily be in an appendix.
Figure 3
- The class colors here and in the other figures are hard to distinguish, could the be more distinct?
- It would be useful to have the same color scale as in Figure 5, etc.
Figure 4.
- Rainfall is discussed multiple times in the text, so rainfall and snowfall should be parsed in panel a.
- The winter period isn’t analyzed, could it be cut out (and possibly included in the Appendices) to make the summer seasons bigger?
- I don’t see any blue or grey shaded areas on my printed version.
- Are there diurnal variations in ice velocity?
Figure 5
- It would be useful to include how to read figure 5 a-c, f-h in the caption including how the curves relate to the bounds to determine behavior.
- Scale the color bars to be the same number of days such that it’s clear that the 2022 data doesn’t go to the end of the melt season and they should be the same across Figures (right now Figure 7 has a different color scale for 2022).
- The vastly different ranges on the x and y axes make it difficult to interpret behavior (see line comment 304). These should be standardized as much as possible, in ways that highlight the main points of the analysis.
Figure 6
- Could the windows be plotted on subpanels b and d?
Figure 7
- See notes about color scale and axes for Figure 5.
Figure 9
How are the melt seasons combined in panel a?
Figure 10
- Where would the borehole sit in the subglacial plan view maps?
References (not already included in the manuscript)
Mair, D., Nienow, P., Willis, I. and Sharp, M.: Spatial patterns of glacier motion during a high-velocity event: Haut Glacier d’Arolla, Switzerland, Journal of Glaciology, 47(156), 9–20, doi:10.3189/172756501781832412, 2001.
Mair, D., Willis, I., Fischer, U. H., Hubbard, B., Nienow, P. and Hubbard, A.: Hydrological controls on patterns of surface, internal and basal motion during three “‘spring events’”: Haut Glacier d’Arolla, Switzerland, Journal of Glaciology, 49(167), 555–567, doi:10.3189/172756503781830467, 2003.
Sharp, M., Richards, K., Willis, I., Arnold, N., Nienow, P., Lawson, W. and Tison, J.-L.: Geometry, bed topography and drainage system structure of the haut glacier d’Arolla, Switzerland, Earth Surf. Process. Landforms, 18(6), 557–571, doi:10.1002/esp.3290180608, 1993.
Citation: https://doi.org/10.5194/egusphere-2023-618-RC2 - AC3: 'Reply on RC2', Coline BOUCHAYER, 25 Nov 2023
Peer review completion
Journal article(s) based on this preprint
Data sets
Dataset at a 3h resolution Coline Bouchayer https://doi.org/10.5281/zenodo.7648444
Model code and software
GitHub repository to process the data, published in Zenodo Coline Bouchayer https://doi.org/10.5281/zenodo.7648470
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 634 | 362 | 36 | 1,032 | 20 | 27 |
- HTML: 634
- PDF: 362
- XML: 36
- Total: 1,032
- BibTeX: 20
- EndNote: 27
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Coline Bouchayer
Ugo Nanni
Pierre-Marie Lefeuvre
John Hulth
Louise Steffensen Schmidt
Jack Kohler
François Renard
Thomas V. Schuler
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4931 KB) - Metadata XML