the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Feb 2024
Widespread increase in discharge from West Antarctic Peninsula glaciers since 2018
Abstract. Many glaciers on the Antarctic Peninsula have retreated and accelerated in recent decades. Here we show that there was a widespread, quasi-synchronous and sustained increase in grounding line discharge from glaciers on the west coast of the Antarctic Peninsula since 2018. Overall, west Antarctic Peninsula discharge trends increased by over a factor of three, from 0.5 Gt/y/decade during 2017 to 2020 up to 1.6 Gt/y/decade in the years following, leading to a grounding line discharge increase of 7 Gt/y (7.4 %) since 2017. The acceleration in discharge was concentrated at glaciers connected to deep, cross-shelf troughs hosting warm ocean waters, and the acceleration occurred during a period of anomalously high subsurface water temperatures on the continental shelf. Given that many of the affected glaciers have retreated over the past several decades in response to ocean warming, thereby highlighting their sensitivity to ocean forcing, we argue that the recent period of anomalously warm water was likely a key driver of the observed acceleration. However, the acceleration also occurred during a time of anomalously high atmospheric temperatures and glacier surface runoff, which could have contributed to speed-up by directly increasing basal water pressure and, by invigorating near-glacier circulation, increasing submarine melt rates. The spatial pattern of glacier acceleration therefore provides an indication of glaciers that are exposed to warm ocean water at depth and/or have active surface-to-bed hydrological connections. Both atmospheric and ocean temperatures in this region and its surroundings are likely to increase further in the coming decades, suggesting that discharge increases may continue and become more widespread.
-
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
(2326 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2326 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-232', Anonymous Referee #1, 01 Apr 2024
Summary: The authors apply change point analysis to an existing ice discharge dataset (created by the lead author) and then compare the results to time series of terminus position and environmental change. ERA5 air temperatures and CTD time series from the Palmer LTER are used to construct records of environmental change. The authors find that discharge increased around 2018, coinciding with both anomalously warm ocean and air temperatures. However, there is variability in the timing and magnitude of discharge change, such that it is difficult to determine the exact cause of the discharge variability. Still, the authors suggest that the widespread increase in discharge that is observed in recent years is an indicator of ongoing and future sensitivity to climate change across the region.
The paper is well written and fairly easy to read, with the exception of a few minor points described below. I appreciate the use of example glaciers but particularly like the map figures for all glaciers in the region (Figure 4) since it clearly shows variability in discharge change that is difficult to decipher from the example glaciers. Most recommended revisions are minor in nature, with the exception of a few points regarding the way that the change in discharge is described and the discussion of the change with respect to environmental forcings.
Major Comments:
- I appreciate the use of change point analysis to identify changes in linear trends in discharge in the dataset because it minimizes user bias in the trend interpretation. That said, I think that the way the trends are discussed is a bit confusing/misleading at times. The time series seems quite short – 2017-2023 – yet the authors interpret the trends over as little as ~2 years to be indicative of longer-term changes. This might not be the authors’ intention but presentation of trends over time as Gt/y/decade implicitly implies that the trends will persist for decadal time scales. Why not present the trends as Gt/y^2 as has been done elsewhere? Additionally, the authors point to the observed variability in discharge trends as an indicator of differences in sensitivity to environmental forcings that is at least partially due to differences in glacier geometry but then conclude that “discharge increases may continue and become more widespread”. The observed variations in discharge change for individual glaciers and the apparent dependence on geometry means that the observed years-long changes in discharge will likely not persist on decadal time scales at individual glaciers or across the entire region in the coming decades. I recommend carefully revising wording to not make to many large leaps in interpretation of discharge change over the coming decades at the individual and regional scales.
- I think the use of example glaciers is really helpful when including such a large sample size in an analysis. That said, upon careful reading of the methods, I started to wonder if the ten glaciers that you focus on are used for more than just demonstration. On lines 81-82 you say they were selected for detailed examination because they have the strong changes in discharge trends. Does that mean the interpretation of “regional” change throughout the rest of the manuscript is entirely based on analysis of those 10 glaciers? Or are those glaciers simply used to emphasize the potential for large change? If you only analyzed data for those ten glaciers, then that needs to be made much more clear throughout the paper because you are not really performing a regionally-representative analysis if you are focusing on glaciers with end-member change.
- Cryohydrologic warming seems like an extremely unlikely cause for the observed changes in discharge. It is true that refreezing meltwater can increase deformation rates but I find it unlikely that it could cause 100s of meters of added deformation each year (inferred from Figure 2). I recommend that you either add estimates of cryohydrologic warming-enhanced flow for other locations from the literature and a comparison with changes in speed along the western AP, or you remove the mechanism as an explanation for enhanced discharge.
Minor Comments:
- I don’t think you need to describe the discharge dataset in great detail because it is already described in a published paper, but a brief description should be included to make it easier for the reader to interpret this paper. This statement is particularly true given that the referenced paper is still in discussion. Some basic details like the locations of the gates, the number of study sites, and any bias estimates or corrections are needed here. For example, what does “all glaciers and basin definitions” on line 70 mean?
- Is there a limit to the number of observations over which the trends in discharge can be calculated? Or did you only limit the time period over which you will accept change points (20 months cut from each end)? How/why did you decide to exclude 20 months in particular? Did you force the trendlines to include full years of data to prevent the trends from becoming amplified if fit to partial years (for example an austral summer minimum, across a full year, to an austral winter maximum)?
- I really like Figure 4 but I’d also love to see histograms or box plots of the trends before and after the change point as well as the timing of the change point. Those figures would make it easier to determine synchronicity/uniformity in the data.
Citation: https://doi.org/10.5194/egusphere-2024-232-RC1 -
AC1: 'Reply on RC1', Benjamin Davison, 23 May 2024
The authors would like to thank the anonymous reviewer for providing a thorough and constructive review of our manuscript. Please find the attached supplement detailing responses to both sets of reviewer comments and a revised version of the manuscript, with our proposed changes highlighted.
-
RC2: 'Comment on egusphere-2024-232', Anonymous Referee #2, 06 Apr 2024
General comments
Davison et al. combined several datasets for glacier discharge, air temperature from ERA5, modeled runoff from RACMO, and manually-delineated terminus positions to identify and analyze a quasi-synchronous increase in discharge for glaciers on the West Antarctic Peninsula since 2018. They used change point analysis to identify the timing and relative magnitude of changes in discharge throughout the region.
Overall, the paper is well written and easy to follow. The authors provide a thoughtful discussion of mechanisms and uncertainties for attributing the regional speed up to ocean and air temperatures, and target this area for future bathymetric / bed mapping for better understanding, particularly related to submarine melt processes. I think this is a valuable contribution to the broader Antarctic and glaciological scientific communities. I have a few technical suggestions for improving the clarity of the text, noted below.
Technical corrections
- L18: “near-glacier circulation” - add “ocean” or “water”
- L61: “Ground line discharge is the mass…” Nitpicky note here: I suggest adding something to indicate the rate or flux of mass. It sounds more like a static variable in this sentence. Or simplify to e.g., “...the rate of mass flowing across the glacier grounding line towards the sea.”
- Figure 1. Nicely laid out figure. In panel (a), I suggest using a sequential, more colorblind friendly colormap for the Speed variable, particularly to better distinguish it from the Speed change colors in panel (b).
- L74: “Change points are…” I find this sentence confusing, can you rephrase? I think by “discharge trend” you mean the linear trend or regression coefficient, but it would help to be more specific.
- Figure 4: Really nice figure! This is very helpful for interpreting your results.
- Figure 6: Consider a different basemap e.g., LIMA or Sentinel-2 images for the map view panels. REMA is very blurry esp. for Wiggins and Bussey glaciers and not particularly useful for interpretation.
- L210: “was”
Citation: https://doi.org/10.5194/egusphere-2024-232-RC2 -
AC2: 'Reply on RC2', Benjamin Davison, 23 May 2024
The authors would like to thank the anonymous reviewer for providing a thorough and constructive review of our manuscript. Please find the attached supplement detailing responses to both sets of reviewer comments and a revised version of the manuscript, with our proposed changes highlighted.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-232', Anonymous Referee #1, 01 Apr 2024
Summary: The authors apply change point analysis to an existing ice discharge dataset (created by the lead author) and then compare the results to time series of terminus position and environmental change. ERA5 air temperatures and CTD time series from the Palmer LTER are used to construct records of environmental change. The authors find that discharge increased around 2018, coinciding with both anomalously warm ocean and air temperatures. However, there is variability in the timing and magnitude of discharge change, such that it is difficult to determine the exact cause of the discharge variability. Still, the authors suggest that the widespread increase in discharge that is observed in recent years is an indicator of ongoing and future sensitivity to climate change across the region.
The paper is well written and fairly easy to read, with the exception of a few minor points described below. I appreciate the use of example glaciers but particularly like the map figures for all glaciers in the region (Figure 4) since it clearly shows variability in discharge change that is difficult to decipher from the example glaciers. Most recommended revisions are minor in nature, with the exception of a few points regarding the way that the change in discharge is described and the discussion of the change with respect to environmental forcings.
Major Comments:
- I appreciate the use of change point analysis to identify changes in linear trends in discharge in the dataset because it minimizes user bias in the trend interpretation. That said, I think that the way the trends are discussed is a bit confusing/misleading at times. The time series seems quite short – 2017-2023 – yet the authors interpret the trends over as little as ~2 years to be indicative of longer-term changes. This might not be the authors’ intention but presentation of trends over time as Gt/y/decade implicitly implies that the trends will persist for decadal time scales. Why not present the trends as Gt/y^2 as has been done elsewhere? Additionally, the authors point to the observed variability in discharge trends as an indicator of differences in sensitivity to environmental forcings that is at least partially due to differences in glacier geometry but then conclude that “discharge increases may continue and become more widespread”. The observed variations in discharge change for individual glaciers and the apparent dependence on geometry means that the observed years-long changes in discharge will likely not persist on decadal time scales at individual glaciers or across the entire region in the coming decades. I recommend carefully revising wording to not make to many large leaps in interpretation of discharge change over the coming decades at the individual and regional scales.
- I think the use of example glaciers is really helpful when including such a large sample size in an analysis. That said, upon careful reading of the methods, I started to wonder if the ten glaciers that you focus on are used for more than just demonstration. On lines 81-82 you say they were selected for detailed examination because they have the strong changes in discharge trends. Does that mean the interpretation of “regional” change throughout the rest of the manuscript is entirely based on analysis of those 10 glaciers? Or are those glaciers simply used to emphasize the potential for large change? If you only analyzed data for those ten glaciers, then that needs to be made much more clear throughout the paper because you are not really performing a regionally-representative analysis if you are focusing on glaciers with end-member change.
- Cryohydrologic warming seems like an extremely unlikely cause for the observed changes in discharge. It is true that refreezing meltwater can increase deformation rates but I find it unlikely that it could cause 100s of meters of added deformation each year (inferred from Figure 2). I recommend that you either add estimates of cryohydrologic warming-enhanced flow for other locations from the literature and a comparison with changes in speed along the western AP, or you remove the mechanism as an explanation for enhanced discharge.
Minor Comments:
- I don’t think you need to describe the discharge dataset in great detail because it is already described in a published paper, but a brief description should be included to make it easier for the reader to interpret this paper. This statement is particularly true given that the referenced paper is still in discussion. Some basic details like the locations of the gates, the number of study sites, and any bias estimates or corrections are needed here. For example, what does “all glaciers and basin definitions” on line 70 mean?
- Is there a limit to the number of observations over which the trends in discharge can be calculated? Or did you only limit the time period over which you will accept change points (20 months cut from each end)? How/why did you decide to exclude 20 months in particular? Did you force the trendlines to include full years of data to prevent the trends from becoming amplified if fit to partial years (for example an austral summer minimum, across a full year, to an austral winter maximum)?
- I really like Figure 4 but I’d also love to see histograms or box plots of the trends before and after the change point as well as the timing of the change point. Those figures would make it easier to determine synchronicity/uniformity in the data.
Citation: https://doi.org/10.5194/egusphere-2024-232-RC1 -
AC1: 'Reply on RC1', Benjamin Davison, 23 May 2024
The authors would like to thank the anonymous reviewer for providing a thorough and constructive review of our manuscript. Please find the attached supplement detailing responses to both sets of reviewer comments and a revised version of the manuscript, with our proposed changes highlighted.
-
RC2: 'Comment on egusphere-2024-232', Anonymous Referee #2, 06 Apr 2024
General comments
Davison et al. combined several datasets for glacier discharge, air temperature from ERA5, modeled runoff from RACMO, and manually-delineated terminus positions to identify and analyze a quasi-synchronous increase in discharge for glaciers on the West Antarctic Peninsula since 2018. They used change point analysis to identify the timing and relative magnitude of changes in discharge throughout the region.
Overall, the paper is well written and easy to follow. The authors provide a thoughtful discussion of mechanisms and uncertainties for attributing the regional speed up to ocean and air temperatures, and target this area for future bathymetric / bed mapping for better understanding, particularly related to submarine melt processes. I think this is a valuable contribution to the broader Antarctic and glaciological scientific communities. I have a few technical suggestions for improving the clarity of the text, noted below.
Technical corrections
- L18: “near-glacier circulation” - add “ocean” or “water”
- L61: “Ground line discharge is the mass…” Nitpicky note here: I suggest adding something to indicate the rate or flux of mass. It sounds more like a static variable in this sentence. Or simplify to e.g., “...the rate of mass flowing across the glacier grounding line towards the sea.”
- Figure 1. Nicely laid out figure. In panel (a), I suggest using a sequential, more colorblind friendly colormap for the Speed variable, particularly to better distinguish it from the Speed change colors in panel (b).
- L74: “Change points are…” I find this sentence confusing, can you rephrase? I think by “discharge trend” you mean the linear trend or regression coefficient, but it would help to be more specific.
- Figure 4: Really nice figure! This is very helpful for interpreting your results.
- Figure 6: Consider a different basemap e.g., LIMA or Sentinel-2 images for the map view panels. REMA is very blurry esp. for Wiggins and Bussey glaciers and not particularly useful for interpretation.
- L210: “was”
Citation: https://doi.org/10.5194/egusphere-2024-232-RC2 -
AC2: 'Reply on RC2', Benjamin Davison, 23 May 2024
The authors would like to thank the anonymous reviewer for providing a thorough and constructive review of our manuscript. Please find the attached supplement detailing responses to both sets of reviewer comments and a revised version of the manuscript, with our proposed changes highlighted.
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 345 | 82 | 27 | 454 | 13 | 17 |
- HTML: 345
- PDF: 82
- XML: 27
- Total: 454
- BibTeX: 13
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
Benjamin J. Davison
Anna E. Hogg
Carlos Moffat
Michael P. Meredith
Benjamin J. Wallis
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2326 KB) - Metadata XML