the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Dec 2025
Firn microstructure variations under extreme metamorphism at Allan Hills Blue Ice Area, Antarctica
Abstract. The spatial variability of firn properties, including microstructure, is critical for mass-balance estimates and ice-core interpretations. This includes local and regional variations in firn microstructure, which influence compaction and air movement in the firn. Some of the oldest ice cores are located in blue-ice areas and originated from low-accumulation areas. Surface processes, including small variations in surface conditions and topography, may cause significant alteration of the firn microstructure and bulk firn properties in low-accumulation areas, as firn may interact with surface processes for long periods of time (e.g., hundreds of years). Here, we examine the impact of very low-accumulation rates (A = O(10-3 - 10-2 m a-1)) on firn microstructure using micro-computed tomography data from a firn core (PICO2303) retrieved in the accumulation zone at Allan Hills Blue Ice Area (BIA), Antarctica during the 2023–2024 NSF-COLDEX field season. First, we analyze microstructural parameters such as total porosity, density, specific surface area, degree of anisotropy, and structure thickness, as well as 2D and 3D reconstructed digital datasets. Subsequently, we compare our observations with a firn core from Dadic et al. (2015) that we refer to here as AHMIF, and located ∼9 km downslope. In the PICO2303 core, we find 1) faceted grains, 2) wind crusts, 3) vertically oriented networks, and 4) specific surface area of 1.5 m2 kg-1 greater and density of 205 kg m-3 lower below the surface compared to those measured in the AHMIF core. We propose that these characteristics are a result of differences in both surface processes (e.g., formation and decay of sastrugi) and subsurface evolution (e.g., depth-hoar formation) that likely occur at varying degrees. We model surface snow density as a function of wind speed, which is approximated from a katabatic wind model. We find modeled surface density closely predict data at AHMIF and PICO2303 with errors of 8.2 % and 0.4 %, respectively, which suggests that core differences are influenced by spatial variability of surface wind and topography. Broadly, the differences between AHMIF and PICO2303 may impact near-surface air movement with resultant implications on post-depositional modification of stable water-isotope ratios and gas exchange.
Competing interests: At least one of the (co-)authors is a member of the editorial board of The Cryosphere.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(17135 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-5040', Martin Schneebeli, 29 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5040/egusphere-2025-5040-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2025-5040-RC1 -
RC2: 'Comment on egusphere-2025-5040', Anonymous Referee #2, 14 Apr 2026
The manuscript (MS) of Horlings et al. compares the microstructure of two sites from the Allen Hills the snow/firn is/was undergoing extreme metamorphism. The two cores compared show in the uppermost 5 m extreme differences in microstructure and density. Therefore this comparison deserves publication.
In its present form the manuscript is lacking a thorough discussion of the processes which may have caused the differences in microstructure and the related physical properties. Therefore I recommend major revisions.
General comments
The differences in the microstructure of the two core sites (Pico2303 and AHMIF published by Dadic et al. (2015)) are obvious in some figures and 3d reconstructions presented, also in SSA and density but the comparison is limited by the low resolution of the AHMIF core. Below 30 cm the AHMIF core is older than 100 years and may be older than 600 years at the bottom in 5 m depth. No ages are given for the Pico2303 core. The poor age constraints of both cores make it difficult to develop ideas over which time scales the microstructures develop, have developed or which processes have worked on them. It seems obvious that microstructures developed over very different time spans and have been undergoing different processes are compared. The authors hardly discuss this aspect. They do not try to provide age estimates.Helpful information how the snow patches at these sites developed during the time of activities between 2011 and 2023 are not given or do not exist.
My impression is that the AHMIF core shows a discontinuity or an hiatus at 30 cm depth. The high SSA values at the surface (top ~7 cm) indicate relatively recent deposition above much 'older' firn.
The katabatic winds in this hilly region certainly cause somehow the observed features and properties. A basic question, however, is not discussed: to which densities can wind pack snow grains? What is the highest packing density of snow grains? The high densities (up to more than 600 kg/m3) observed close to the surface in the AHMIF core are exceptional and therefore interesting but if a density of 550 kg/m3 for snow is assumed as the highest packing density (densification stage I) then densities above 550 kg/m3 may not be anymore the result of wind packing. Internal deformation is commonly invoked in explaining densities above 550 kg/m3 (stage II) observed in the AHMIF core already as shallow as ~30 cm depth. Is it possible that the AHMIF site was/is undergoing significant erosion or sublimation or did it slowly move from an accumulation region into an ablation site with higher winds and significantly more erosion/sublimation? Blue ice is indicated not far below the AHMIF position in the satellite picture (Figure 2). The AHMIF site moves ~16 cm/year towards a blue ice region where more sublimation/erosion must occur. Where in the AHMIF snow patch is accumulation predominant and where ablation. Is this known? May a possibility to estimate ages.
Spaulding et al., 2012 in their conclusions:... Blue ice and adjacent firn areas are thinning at an average rate of 1.5±1 cm per year. Elsewhere they write that the snow cover is known to change seasonally and annually and sastrugi up to 1 m height are not rare. The focus on wind packing as the only explanation appears to be a very restricted view and explanation.
I am wondering if the Herron-Langway firn model can't be used to derive some estimates of the ages of the cores. Firnification stage I does not depend on accumulation but the densification rate does depend on the accumulation rate. Densities of ~600 kg/m3 are generally observed below 15 m depth in firn for annual mean temperatures of -30°C. A question then is if wind is able to erode firn of densities above 550 kg/m3 or if it is predominantly sublimation. Katabatic winds are always dry winds. If thinning (1.5±1 cm/a) means sublimation then 1 m of firn is removed in about 100 years at the AHMIF site. If such a scenario is plausible then the two exceptional microstructures may not be so exceptional anymore to describe the differences in the observed evolution of the microstructure, density, SSA and grain growth.Helpful for the reader would be if the information about the core sites would not be so widely distributed and a short summary of the Spaulding et al. paper would be nice to read somewhere in the introduction section of the MS. The maps in the Spaulding- paper seem clearer to me to explain and better understand the differences of the two sites.
Abstract
L3:
Some of the oldest ice cores are located in blue-ice areas ...
May be instead: Some of the ice cores with the oldest ice are located in blue-ice areas.L6:
.. impact of very low-accumulation rates (A= O(10−3 −10−2 m a-1))
is estimated or measured?
Easier to read would be: 1 mm to 1 cm per year or so, ice or water equivalent or snowL12ff
Differences between parameters of the two cores alone are not very helpful if their averages are not given.
-------------1.1 Core LocationFigure 2: units in km would be easier to read
L91 and L94/95: -76.667°S, 159.25°E, ... missing is °S/E
Somewhere within in the introduction section I am missing useful site information about this blue ice region given by Spaulding et al., 2012.
-------- 1.2 Climate at Allan Hills BIA
Missing here is the annual mean temperature or a 10 m firn temperature and summer mean and not only daily variationsFigure 4:
- Position of the AWS? how close was it to the Pico23 site? Please mark it in Fig. 2
- which data are presented? hourly or daily averages ?
- wind was measured at which height?
- is/was it a permanent AWS and if yes then please also present annual data
- please define grid north - is it identical to true north?
- there are obviously no accumulation sticks at this site although revisited over 4 field seasons
-------- 1.3 Micro-computed tomography (micro-CT)L119:
rho = (m3/m3)/kg/m3 = m3/kg ??? change "/" to "*"
L120
DA = 1− (λmin/λmax)
this definition does not allow to infer if the preferred alignment of a structure is vertical or horizontal
Please write somewhere how many µCT samples did you take or what resolution you have chosen?-------- 1.4 Comparison of cores
Figure 5:
most likely shown are vertical projection?
Helpful if noted: scale bar is 2 mm--------- 2 Results
L170:
There is some trend in DA obvious despite the high scatter
Wind crusts:
They take a lot of space in the MS in particular as their formation is not at all fully understood. They are nice visible features to have. The wind crusts I have seen did not show large grains (L196: large-grained WC) and not much is known about their spatial distribution. They are always restricted to a very limited area (dimension of a snow patch, order of 10 m by 10-50 m or so). So it might also be statistical problem to observe or not observe a wind crust.
------- 3.1 Operating processesFigure 7:
In Fig. 7e,f Structure thickness and SSA show a change in their gradients below about 4 m depth. Changes of the slope are not observed in DA and density. This is not discussed.
Figure 8:
- The best-fit of the structure thickness in Fig. 8a seems to be the best fit of the two data sets. Is this meaningful and why?
- You shift the best-fit in SSA of the AHMIF core into the Pico2303 core SSA. This is not discussed why.Snow patches in blue ice regions belong probably to the regions with the most dynamic snow surfaces in any respect - in particular Sastrugi up 1 m height are reported by Spaulding et al., 2012. I am not wondering why wind crusts are inclined or horizontal and why temperature gradient metamorphism above densities of 350 kg/m3 should not be possible. TGM is commonly observed in areas with "normal" accumulation rates within limited boundaries in the controlling parameters. Snow on blue ice is not a standard case.
L190
why is higher SSA indicating high winds in the top most centimeters but shows the opposite relationship found in greater depth where SSA decreases with increasing density? How do you explain?
Instead of showing as inset the SSA versus depth profile I suggest to show a scatter plot SSA vs. density. I assume that all data are derived from the same samples. Therefore it would be interesting to see scatter plots of the different parameters if they provide additional information.
L192:
Lack of wind crusts below ~30 cm is probably not surprising as the firn is old and the metamorphism phase of the AHMIF core is unknown. Do wind crusts survive heavy metamorphism?
L200±
Air/vapour movement in porous firn is an interesting aspect for the interpretation of isotope profiles but challenging in a blue ice region when the age is not well constrained. So here more a side aspect here.
L212: Effect of wind
see my comments about wind packing density above
L247
"... while higher wind speeds generate high-impact deposition, and thus the higher density cap and higher SSA near the surface."
I do not understand.
Recomentation
The MS presents another interesting new data of microstructure and physical properties from a blue ice region studied not much yet. However, the discussion and the interpretation appears too focussed to a single explanation: wind packing. Other interpretations appear possible. Therefore, I recommend major revision.Citation: https://doi.org/10.5194/egusphere-2025-5040-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 237 | 491 | 37 | 765 | 43 | 57 |
- HTML: 237
- PDF: 491
- XML: 37
- Total: 765
- BibTeX: 43
- EndNote: 57
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1