Preservation of Glaciochemical Signals in a Warming Tropical Ice Cap: Evidence from the 2022 Quelccaya Ice Core

Gomes Ilha, João; Brum Poitevin Portella, Manoela; Vilca Gomez, Oscar; Gaudie Lindau, Filipe; Potocki, Mariusz; Barbaro, Elena; Kaspari, Susan; Barbante, Carlo; Cardia Simões, Jefferson; Mayewski, Paul Andrew

doi:10.5194/egusphere-2026-4779

Preprints

https://doi.org/10.5194/egusphere-2026-4779

Preprints

23 Sep 2026

| 23 Sep 2026

Status: this preprint is open for discussion and under review for The Cryosphere (TC).

Preservation of Glaciochemical Signals in a Warming Tropical Ice Cap: Evidence from the 2022 Quelccaya Ice Core

João Gomes Ilha, Manoela Brum Poitevin Portella, Oscar Vilca Gomez, Filipe Gaudie Lindau, Mariusz Potocki, Elena Barbaro, Susan Kaspari, Carlo Barbante, Jefferson Cardia Simões, and Paul Andrew Mayewski

Abstract. Paleoclimate records stored in ice cores are facing great threat as on-going mass loss is rapidly accelerating. To study Amazon’s paleoclimate, an ice core was drilled in the Quelccaya ice cap, in Peru, in 2022. However, as a first step, post-depositional alterations need to be discussed and described. During drilling, one water table (WT) associated with the firn-ice transition and a major fracture at 45 m depth were identified. Here we present the first chemical results retrieved from the ice core and discuss the preservation state of the record in terms of post-depositional alterations. Concentrations of levoglucosan, Na⁺, K⁺, Cl⁻, SO₄²⁻ and NO₃⁻ were analyzed and compared with the relatively insoluble and immobile black carbon signal. Furthermore, signal preservation was evaluated using descriptive statistics, covariance analysis, ionic balance, and multivariate classification. Statistical analyses reveal enhanced ion concentration atop the WT. The section below the WT extending to the fracture is marked by a highly altered chemical signal, as indicated by multivariate analysis. Within this interval, Cl⁻ is strongly depleted and Na⁺ shows increased association with SO₄²⁻ and NO₃⁻, while the ionic balance indicates increased acidity relative to the WT onset. Below the fracture, chemical signals are attenuated but largely preserved. Principal Component Analysis coupled to Fuzzy C-Means clustering independently identifies the WT, altered ice interval, the fracture zone and a deep ice anomaly, suggesting variable preservation along the core. We define depth intervals characterized by preserved, amplified, altered and attenuated signals and show that despite significant meltwater influence, post-depositional alterations are locally concentrated rather than affecting the entire record. These results provide a basis for future paleoclimatic investigations using what could be the last Quelccaya Ice Cap ice core before continued melting irreversibly alters the record.

Received: 11 Aug 2026 – Discussion started: 23 Sep 2026

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.

Download & links

Preprint (PDF, 2230 KB)

Supplement (559 KB)

Download & links

Status: open (until 04 Nov 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-4779', Anonymous Referee #1, 22 Oct 2026 reply
Peer Review of Ilha et al.: “Preservation of Glaciochemical Signals in a Warming Tropical Ice Cap: Evidence from the 2022 Quelccaya Ice Core”
In this manuscript, Ilha et al. present ice core impurity measurements from a new ice core drilled from the Quelccaya Ice Cap in 2022. Like many alpine glaciers, especially in tropical regions, the Quelccaya Ice Cap is threatened and its climate archive has become compromised due to widespread surface melt. It is well known that substantial meltwater generation in the firn column can alter climate signals contained in the glaciochemistry of the ice column. Therefore, this paper represents an attempt to unravel the effects of meltwater on the climate signal.
Although this dataset has clear potential and a study of this sort will be necessary for interpreting the climate signals in the core, this work suffers from persistent issues ranging from incorrect or underexplained methodological choices to vague and incomplete interpretations, which I elaborate upon below. I therefore cannot recommend its publication in its current form in The Cryosphere.
Main points
The data presented here were not adequately placed in the context of previous studies on the Quelccaya Ice Cap. They note that two previous ice cores were drilled there in 1983 and 2003, but not that a surface-to-bedrock core was also drilled in 2018 (Clifford et al., 2023 — this study was cited but receives limited discussion). Numerous studies have examined these cores and the glaciological conditions at the ice cap.

Thompson et al. (2013) report major ions from the surface-to-bedrock 2003 core and Clifford et al. (2023) report a range of geochemistry including major ions from the 2018 core, yet no attempt was made in this study to compare to these records. For instance, had they plotted the data in this core on time (presuming a chronology can be developed by counting annual layers in black carbon), do they see any agreement with the previous records below the fracture where they claim climate signals are preserved?

More generally, do statistics of impurity concentrations and variability compare favorably to those reported in the 2003 and 2018 cores? It is difficult to gauge this because they report impurities in the more unusual nmol L-1, whereas most other ice core studies including Thompson and Clifford equivalently use µg L-1 or ppb.

Was this core drilled at the same location as any of the previous cores? The two 2003 cores were drilled almost 2 km apart, so which site on the ice cap was this core from?

The discussion of the water table and attenuated climate signals would benefit from acknowledging earlier observations. Firn temperature measurements in the mid-late 1970s show that the firn column was isothermal at 0°C below ~5 m (Thompson et al., 1980). The same study stated that ice lenses were visible in the firn column implying summer surface melt, although they suggested these did not greatly affect the seasonal signal. Later, the 2003 core was found to already have significantly altered climate signals compared to the 1983 core (Thompson et al., 2011). During drilling in 2018, Clifford et al. (2023) report meltwater buildup in the borehole implying the existence of a water table around 23 m depth, similar though slightly deeper than reported in this study. Surface melt has therefore been a persistent feature at this site for at least five decades, and almost certainly longer. This study would be greatly improved if the melt effects they observe were discussed in the context of prior work. Has it gotten worse? Does the melt penetrate deeper or has the water table expanded? Does it affect more geochemistry than in the past? What implications do decades of warmth and summer melt have on the authors’ claim of retained climate signals below the fracture?

The manuscript needs to articulate what the 2022 record establishes beyond what was reported in previous observations of meltwater and chemical alteration at Quelccaya.

It is unclear to me what a “fracture” physically describes. The main observation provided is that the drill dropped by approximately 1 m. If it were a buried crevasse or a melt channel, then there would be no ice from that 1-m depth range, yet they report chemistry from the fracture, so presumably there was no gap in the ice core. What then is the “fracture”, what caused the drill to drop, and how does this relate to the melt alteration?

Converted to ppb (= µg L-1), the y-axes in Figure 2 for Na, Cl, SO4, NO3, and K reach 690, 638, 432, 465, and 938 ppb, respectively. These are very high concentrations. Is this expected? Did they replicate any analyses on adjacent core sections? Are these measurements similar to the ranges reported by Clifford et al. for the 2018 core? Glancing at that dataset, the measurements in this paper seem broadly higher although I did not do a rigorous analysis because the data from this study are not yet publicly available. I would have liked to see a discussion of the high concentrations.

The description of field storage leaves some questions about how the cores were protected from melting. In Section 2.1, it is stated that the temperature inside the borehole was 0°C (I note that typically the temperature in the top few meters is not 0°C even if the ice column is isothermal) and that they stored ice cores in insulated boxes that they buried in a snow trench. If the ice column was truly 0°C, how did they ensure the boxes stayed cold during the week of drilling? They mentioned that the temperature was logged and no melting was recorded, but I would have liked to see this quantified, with a statement like “the maximum temperature reached was __ °C”. Because the study concerns melt-induced redistribution, demonstrating that handling and storage did not introduce comparable alterations is particularly important.

The Cl/Na ratio in Fig 3 has unusual step-changes that are not discussed. These occur every ~5 meters between 20 and 50 m. Broad shifts in this ratio are attributed in the study to the water table, but this odd step-change pattern is not mentioned. It would be worthwhile for the authors to investigate this, and confirm that these shifts don’t correspond to artifacts such as IC analysis runs (or, more concerningly, groupings of cores in ice core boxes).

If the desire is for a positive ionic balance to indicate acidity, it should be calculated by subtracting cations from anions, however the y-axis label of Figure 9 says “Cations - anions”. Lines 419-420 correctly say “When calculating the ionic balance by subtracting major cations from major anions…”, so there is an inconsistency and it is unclear which was used. If the y-axis label is correct, then their interpretation of acidity needs to be completely reworked or removed. I also encourage them to check and clarify in the text that they used charge-weighted concentrations. I also wonder why Figure 9 was cut off at 50 m; it would be useful to see if the anomalous ionic balance below 40 m is unique in the core or if deeper sections show this.

The quantity described as signal-to-noise ratio (SNR) seems flawed and does not appear to establish signal preservation in the way it is interpreted here. It is calculated by subtracting from a sample concentration the median concentrations in “background ice” (defined as ice below 50 m) and dividing this difference by the median absolute deviation of the “background”. Therefore, any interval whose concentrations are higher than the “background” would be interpreted as having a high SNR despite possibly containing no seasonal or climatic information; conversely an interval with low concentrations but well-resolved seasonal cycles would be considered as having a low SNR. In effect, the SNR is just standardized concentration. This formula therefore seems unfit for their purpose and the SNRs are misinterpreted throughout.

Moreover, heightened variability is frequently interpreted as evidence of signal preservation, which is a flawed assumption. Meltwater redistribution can leave a variable chemical profile while altering the timing, amplitude, and composition of depositional peaks, and may even increase variability through localized enrichment. Variability alone therefore does not establish preservation of the original depositional record.

The pronounced Cl depletion is an important observation, but the inference that Cl elutes faster than other ions is contrary to existing literature and is not established by concentration ratios alone. These ratios reflect both initial deposition and subsequent redistribution, and the proposed acidity mechanism remains insufficiently supported (e.g., line 418). Without a physical mechanism for what is driving Cl elution, I am not convinced based on this work that the Cl/Na ratio is a reasonable reconstruction of Cl elution speed. It would be particularly useful to compare not just Cl to Na, but to all the other major ions which should, in theory, elute at different rates.

On this note, I believe the interpretation of ion correlations needs adjustment. Sections 3.6 and 3.7 interpret changes in ion correlations as evidence of new chemical associations (described imprecisely as “bonds”), but differential elution can produce these changes without forming new bonds. Decoupling of Na+ and Cl– does not establish that chloride becomes associated with H+, nor does increased Na+–sulfate/nitrate covariance demonstrate formation of those salts. These changes in covariance could result from differential elution without any change in preferred chemical association.

The contribution of the PCA and clustering analysis in Section 3.8 is unclear. It seems like this analysis simply shows that the chemistry in the water table is different from outside of it, yet this was already shown in previous sections. It’s unclear to me what this statistical analysis adds. If the authors choose to keep it, I suggest they fit the PCA to the full core and not just the upper 45 m.

Other points
In addition to imprecise and wordy writing, the paper suffered from pervasive structural issues. Just to give an example, lines 219-225 in Section 3.1 discuss the Cl/Na ratio but that is the subject of Section 3.2. Section 3.2 discusses the water table, yet ends with 3 sentences on the data below the water table, which is the subject of Section 3.3. Sections 3.4, meanwhile, returns to the subject of the water table ionic alteration. The entire paper should be revised for clarity and structure.

Given the very high major ion concentrations it would be useful to plot these on a logarithmic scale in Fig 2 and elsewhere, as was done for BC. (I note that plotting BC alone on a logarithmic scale is a misleading comparison with the linear scale of the major ions.)

The levoglucosan data (Fig. 2, Table S2) is virtually all under the limit of quantification, and Table S2 shows that mean concentrations for all depth ranges except 0–20 m are under the method detection limit. What is the rationale for reporting it?

I do not believe this study demonstrates that BC is not affected by melt percolation as the authors claim (Lines 202-203, 216-218). Besides the misleading y-axis plotting (see above), their argument is based on the fact that apparent seasonality is retained because it is relatively insoluble. It would be worthwhile, then, to also show dust measurements. Dust is also largely insoluble and so should show retained variability and seasonality similarly to the BC. To test whether signals are preserved, one could test whether the seasonal cycles of dust and BC both remain stable through time, yet this is not done.

I was surprised the authors did not cite Huber et al. (2024) which discusses signal alteration in a melting glacier archive in Switzerland. It would be useful to date the 2022 Quelccaya core in this study to make a figure similar to Fig. 2 of Huber, which compares geochemistry from cores at the same site drilled years apart.

Minor comments
Table S1: The units are incorrect for BC

They report summary statistics on geochemistry that is not shown (e.g., Ca, Mg, MSA, Br). It would be useful to show these on depth in a supplementary figure.

Lines 36-39: Here and elsewhere, the citation lists could be more selective, with more focused discussion of the studies most relevant to the argument.

Line 73: the sampling resolution is not unprecedented. Many other ice core sites have been sampled at a high resolution.

Figure 1: The Yellow frame of the Amazon basin is very difficult to see. South America does not need to be in the legend.

Line 216: I believe this should be Supplementary Table 2?

Lines 231-236: I do not understand their discussion of local sources of Cl and Na. They mention “ecosystems” that may influence the ice core but don’t provide any actual physical sources for non-marine Na or Cl emission and transport to the ice core site. Likewise, deep convection is mentioned but its relevance is not explained.

Line 251: The Cl/Na ratio does not look close to the sea-spray ratio below 25 m. It has step-change variability. It would be useful to plot the sea-spray ratio as a dashed horizontal line on that panel.

Line 260: what does “biomass production” mean?

Lines 263-266: The discussion of local sources of NH4 and Cl is very weak. I believe Thompson et al. (2013) attributed NH4 variability to Amazon biological conditions responding to precipitation variability. This is never explained here, and seems in direct conflict with coincident Cl enrichment from hyperarid environments of the central andes.

Lines 266-268: I believe this is the first mention of possible elution under the fracture. This seems an important point and is worthy of discussion (see above for my suggestion to discuss the history of warm firn at this site).

Lines 283-293: The proposed sulfate sources require further justification. First of all, they are discussing elevated sulfate in the top 4 m as a result of an eruption that is stated to have lasted from 2016–2020 (although its temporal variability is not mentioned). Glancing at black carbon in Fig. 5, I’d estimate that the top 4 m contain 2020, 2021, and the start of 2022 until drilling. Therefore, the eruption cannot easily account for the elevated sulfate for the whole layer. Secondly, it is stated that sulfate could have a biomass burning origin (Lines 289-290); to my knowledge, this would be a novel interpretation of ice core sulfate. Thirdly, it is stated that the lack of an increase in nitrate implies that the sulfate is not anthropogenic (Lines 290-291); this inference requires reconsideration because studies from Greenland and elsewhere show that anthropogenic nitrate and sulfate have very different atmospheric transport processes and therefore their ice core trends are very different. Lastly, it is stated that elevated levoglucosan implies higher biomass burning; how do the authors account for low black carbon during this interval?

Lines 320-324: This would be a great opportunity to compare the elevated impurities in deep ice in this core to other Quelccaya cores.

Figure 6 would benefit by adding a line that is the normal sea-salt ratio

Lines 362-366: I was under the impression that Cl was also positioned at grain boundaries but I am not very familiar with that literature. Their statement that it is inside grains (it would be good to cite a study showing this) seems in conflict with the next sentence that says it is easily “transported in the very fine grain limits”.

Figure 7 seems like it adds little to what Figure 6 already showed

Figure 10: Swap x and y axes of panel b so PC2 is on y-axis in both

The conclusions could be shortened to 1–2 paragraphs focused on the principal findings

There are typographical errors throughout that warrant a close editorial pass. Additionally, some of the language seemed unusual for a scientific paper, for example at Line 67 they describe previous cores having been drilled by “Thompson’s team”.

References
Clifford, H. M., Potocki, M., Rodda, C., Dixon, D., Birkel, S., Handley, M., et al. (2023). Prefacing unexplored archives from Central Andean surface-to-bedrock ice cores through a multifaceted investigation of regional firn and ice core glaciochemistry. Journal of Glaciology, 69(276), 693–707. https://doi.org/10.1017/jog.2022.91
Huber, C. J., Eichler, A., Mattea, E., Brütsch, S., Jenk, T. M., Gabrieli, J., et al. (2024). High-altitude glacier archives lost due to climate change-related melting. Nature Geoscience, 17(2), 110–113. https://doi.org/10.1038/s41561-023-01366-1
Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Zagorodnov, V. S., Howat, I. M., Mikhalenko, V. N., & Lin, P.-N. (2013). Annually Resolved Ice Core Records of Tropical Climate Variability over the Past ~1800 Years. Science, 340(6135), 945–950. https://doi.org/10.1126/science.1234210
Thompson, Lonnie G. (1980). Glaciological Investigations of the Tropical Quelccaya Ice Cap, Peru. Journal of Glaciology, 25(91), 69–84. https://doi.org/10.3189/S0022143000010297
Thompson, Lonnie G., Mosley-Thompson, E., Davis, M. E., & Brecher, H. H. (2011). Tropical glaciers, recorders and indicators of climate change, are disappearing globally. Annals of Glaciology, 52(59), 23–34. https://doi.org/10.3189/172756411799096231

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4779-RC1
RC2: 'Comment on egusphere-2026-4779', Anonymous Referee #2, 25 Oct 2026 reply

Review of the manuscript “Preservation of Glaciochemical Signals in a Warming Tropical Ice Cap: Evidence from the 2022 Quelccaya Ice Core”

General comments

This manuscript presents the first glaciochemical results from the 2022 Quelccaya ice core, focusing on the impact of post-depositional processes on the preservation of chemical signals. The topic is relevant, particularly considering the increasing vulnerability of tropical ice-core archives to warming and meltwater percolation. The availability of a new deep ice core from Quelccaya represents an opportunity to investigate these processes and assess the paleoclimatic potential of the remaining ice archive.

However, I believe the manuscript requires substantial revision before its conclusions can be adequately supported. My main concern is that, despite the large amount of analytical data and the numerous statistical approaches employed, the interpretation remains fragmented and often insufficiently supported by quantitative evidence. Several conclusions concerning signal preservation, ion mobility, and the influence of specific glaciological features appear to rely primarily on qualitative observations, without a sufficiently systematic evaluation of the chemical records.

Below the main issues that should be addressed:

1) Abstract

The abstract would benefit from substantial restructuring. Although the main elements of the study are present, they are not organized into a clear narrative. In particular, the research question and the specific objectives are not explicit, making it difficult to understand from the outset what the authors intend to establish.

The opening sentences move from the ongoing disappearance of tropical ice-core archives to Amazonian paleoclimate reconstruction and post-depositional alterations without adequately explaining how these topics relate to the central objective of the paper. Similarly, the results are presented mainly as a sequence of individual observations and statistical findings, without a clear distinction between the main findings and supporting details.

I suggest reorganizing the abstract around a more straightforward structure: the scientific problem being addressed, the analytical approach, the key results, and their broader implications. The emphasis should be placed on the scientific interpretation rather than on listing individual observations and statistical procedures.

2) Statistical analysis and quantitative support for the conclusions

One of the main weaknesses of the manuscript concerns the relationship between the statistical analyses and the scientific conclusions.

On the one hand, the authors apply numerous statistical approaches, including measures of variability, correlation analyses, and multivariate techniques. On the other hand, several central claims are not supported by adequate statistical testing.

For example, statements concerning increasing or decreasing ionic concentrations across different sections of the core, changes in chemical behavior at specific depths, and differences between ice above and below the water table are often based on qualitative comparisons. These observations are fundamental to the interpretation of post-depositional processes and should be supported by appropriate quantitative analyses.

I would encourage the authors to adopt a more systematic approach, possibly starting with the identification of significant changes or breakpoints in the chemical depth profiles. This could help establish which species show similar behavior, which ones respond differently, and whether changes occur consistently at specific depths. Such an analysis would provide a more objective basis for the subsequent interpretation of the processes affecting the chemical records.

The same limitation is particularly evident in Section 3.3, where the discussion focuses on a few selected events involving chloride, ammonium, and levoglucosan. Although these examples may be interesting, the interpretation remains essentially qualitative and is based on a very limited portion of an ice core extending over more than 100 m. If the authors intend to investigate relationships among these proxies, I would expect a systematic analysis of their behavior throughout the available record, supported by quantitative indicators rather than a few selected examples.
3) Interpretation of the water table, firn–ice transition, and fracture

The water table (WT) and the fracture identified during drilling represent two central elements of the manuscript, as much of the interpretation is organized around their occurrence.

I agree that the WT may strongly influence ionic concentrations and their vertical distribution. However, the discussion does not adequately consider its relationship with the firn–ice transition. In glacier environments, the occurrence and position of a water table are closely linked to the permeability structure of the firn and to changes in pore connectivity near the firn–ice transition. These aspects should be discussed together.

The interpretation of the fracture is even less convincing. While the authors propose some possible mechanisms to explain the influence of the WT, the physical processes through which the fracture might affect ionic concentrations and variability are not clearly described. It is not evident whether the fracture acted as a preferential pathway for meltwater, whether it was hydraulically connected to other parts of the glacier, or whether its association with changes in chemical composition is simply coincidental.

Before attributing chemical anomalies to the fracture, the authors should provide a clearer physical explanation and supporting evidence for the proposed mechanism.

More generally, the discussion concentrates heavily on the two specific depths corresponding to the WT and the fracture, while the chemical records along the entire core receive comparatively little attention. A more comprehensive examination of their depth-dependent behavior, including broad concentration trends and changes in variability, would considerably strengthen the study.
4) Depth scale, chronology, and comparison with previous Quelccaya ice cores

I strongly recommend considering the use of meters of water equivalent (m w.e.) instead of depth in meters, particularly because the investigated sequence includes both firn and glacier ice.

For example, Figure 3 shows an apparent change in the resolution of BC oscillations around the firn–ice transition. At least part of this change might reflect differences in density rather than an actual modification of the temporal resolution or preservation of the signal. Expressing depth in water equivalent would facilitate a more meaningful comparison between the different sections of the core, provided that suitable density measurements are available.

Another important limitation is the complete absence of a chronological framework, even a preliminary one. The authors repeatedly emphasize that the BC record preserves a clear seasonal signal. If such seasonality is sufficiently recognizable, I would encourage the authors to explore the possibility of developing a preliminary chronology, at least for the upper section of the core where annual variability appears to be identifiable. This would provide an important additional constraint on the interpretation of signal preservation.

Furthermore, the manuscript does not provide a meaningful comparison with the numerous ice cores previously recovered from Quelccaya. I consider this a particularly important omission. Since one of the main objectives is to assess the deterioration of glaciochemical signals under increasing meltwater influence, historical Quelccaya records could provide a valuable reference for evaluating the chemical characteristics of the 2022 core.

Even a preliminary comparison of absolute ionic concentrations, concentration ratios, and signal variability with previously published records could help distinguish post-depositional changes from differences related to atmospheric deposition or natural variability. Such comparisons would require appropriate consideration of chronological coverage, sampling resolution, and analytical differences, but they could substantially strengthen the conclusions.
5) Analytical reliability of low-concentration records

An additional issue concerns the apparent step-like changes visible in the low-concentration portions of some chemical profiles, particularly chloride and, consequently, the Cl⁻/Na⁺ ratio in Figure 3.

From my experience, similar discontinuities can sometimes arise from differences in detection limits, analytical sensitivity, or data processing between measurement batches. I cannot determine whether this is the case here, but the possibility should be investigated and explicitly addressed.

The authors should clarify how concentrations close to or below the detection limits were handled, whether different analytical batches were involved, and whether such effects might influence the observed chemical patterns.

This issue is particularly relevant because some of the interpreted anomalies, including the apparent loss of correlation between Na⁺ and Cl⁻ in the 30–40 m interval, occur in sections where these discontinuities seem especially pronounced.
6) Coherence of the geochemical interpretation

The discussion is fragmented and, in several instances, the interpretation of chemical changes appears insufficiently consistent across the different analyses.

For example, changes in concentration variance or signal-to-noise ratios are sometimes interpreted as evidence of preferential ion elution without adequately considering other available indicators, particularly the Cl⁻/Na⁺ ratio. In some sections, the authors suggest that chloride is more strongly affected by elution than sodium, but the evidence supporting this conclusion is not convincing to me.

I would encourage the authors to reconsider these interpretations by integrating absolute concentrations, concentration ratios, correlations, and changes in variability within a single, coherent framework. In particular, the differential behavior of Na⁺ and Cl⁻ should be assessed consistently throughout the manuscript.

I also find the discussion based on acid–base reactions difficult to follow. Before developing relatively detailed hypotheses concerning chemical reactions and changes in ionic associations, the authors should establish a clearer understanding of the overall behavior of the individual species.

A comparison with previous Quelccaya records, as suggested above, could provide an important starting point. Although differences in absolute concentrations cannot by themselves establish the extent of post-depositional elution, they could help identify potentially depleted or enriched species and guide a more robust interpretation.
At present, I have the impression that the manuscript explores relatively complex geochemical mechanisms without first establishing a sufficiently solid description of the primary chemical evidence.

Overall, I believe the study could be significantly improved by simplifying and reorganizing the discussion, strengthening the quantitative assessment of chemical changes, and developing a clearer link between the observed patterns and the proposed post-depositional mechanisms.
Specific comments

Line18: “However, as a first step, post-depositional alterations need to be discussed and described.” Not very usual to find such a passage in an abstract. I would remove this.

Line19: “fracture”? Fracture of what? Do you mean a crevasse? Please explain better

Line 37-39: is it really needed to put 14 citations here? I think two could be enough, one for deep polar ice cores, one for mountain ice cores. In general, I repeatedly see 4-5 references cited to support a single scientific statement. The number of references of this paper is absolutely exaggerated now.

Line 41-52: this part could be removed or drastically shortened. The authors just need to report two lined to mark the difference between polar and non-polar ice cores. This is not a review, it is not needed to go such in detail.

Line 54: “the region” what region?

Line 57: “photochemical degradation of elements” elements, by definition, cannot undergo degradation or chemical transformation.

Line60-62: this sentence is grammatically awkward and its meaning is unclear, particularly in the second half ("the complete erasure with the disappearance of some glaciers"). I suggest rephrasing.

Line63-65: not clear the link between the position of the record and the ability to record highly resolved information.

Line67: again, 3 lines of citation to report about drilling activities on Quelccaya may be a bit exaggerated.

Line77-86: this final paragraph of the introduction is too long and redundant, particularly in the repeated statements concerning the assessment of signal preservation. I suggest substantially shortening.

Line 95: “The temperature inside the borehole was 0 ° C.” I suggest adding something like “revelaing the temperate condition of the ice cap”.

Line96: “and the base is cold” you have mentioned, just one line above, that borehole temperature was 0°. So, how is it possible to state that the base is cold? This is contradictory. In general the meaning of the passage “and the base is cold; therefore, horizontal movement is negligible as most of the flow happens vertically, compressing the ice” is not very supported. If the ice mass lies on a horizontal bedrock, the absence of vertical ice movement does not depend on the thermal regime of the ice, but only geometrical constrains.

Line100: how did you identify the WT during drilling? What evidences?

Line103: “fracture aquifer” never heard about that. I would simply remove this part. The occurrence of a WT just above the firn/ice transition is something common and compatible with the current glaciological conditions of Quellcaya. But I would remove the discussion about the potential exitance of a fracture aquifer. This is not something well defined in glaciology.

Figure1: pleas add the glacier limit, possibly updated to recent years. Now it is not possible to identify the extent of the ice cap.

Line112: “their laboratory” I suggest to change to something like to the laboratory available there

Line116-117: “Additionally, parts of the archive section are being used for laser ablation (in progress)” by definition a part of the core defined as “the archive” cannot be used for analyses but is taken apart for long-term storage. I suggest to change this part.

Line160: maybe something missing?

Line159-190: this is also a section that would greatly benefit from a substantial reduction. All tools and algorithms applied are well known since decades. The authors does not need to describe their functioning and putting references for them. This is only needed when a new approach is proposed. But this does not seem the case.

Line194-195: “All species show variation in concentration with depth, indicating at least a partially maintained natural variability, i.e., a partially preserved signal.” Showing some variability does not automatically imply that some signal is preserved. I suggest to be less assertive here.

Line202: “compaction”, ice is almost completely uncompressible. This thinning, not compaction. Firn and snow can compact, not glacier ice.

Line205: I see why the WT should influence chemical composition of the ice. But what about the fracture? From what you write, it seems that you were expecting some effect of the fracture on the composition of the ice. Why is that?

Line218: “Hence, it is here considered a local marker of signal preservation.” This claim is not very meaningful to me. What the BC record is telling, is just that BC itself is more resistant to meltwater-related postdepositional processes. It is not a “marker” of preservation. What does it mean?

Line222-223: “or a Cl− enrichment that could be related to elution processes breaking the NaCl molecule" this is chemically incorrect. NaCl is a ionic compound, not a molecule.

Line246: you could cite Baccolo2026 here, where the effect of WT on ionic records is extensively described.

Line265: “produced by local geochemistry”, what does it mean? Local geochemistry of what? No clear

Line274: “as guides”

Line329-330: “Most of the works describing elution in glaciers points out that Cl− behaves similarly to Na+ and is generally the least mobile of major ions” Not true at all, the fact that the Cl-/Na+ ratio can be used as a proxy for melting, depends directly on the differential behavior of the two ions when meltwater is present.

Line338: “Cations and anions bond to form molecules” Please be careful, again. Ionic salts cannot be defined as molecules, this is basic chemistry.

Figure6: in my opinion it would be more informative to display the trend of spearman correlation over depth, calculating it with a mobile window. The current graph is also functional to this, but the information is less evident.

Lin364-366. “However, Cl− is also regarded as suffering more from diffusion in consolidated ice, as in the mid- and deep ice, because it is easily transported in the very fine grain limits due to its smaller size compared to Na+.” Please provide support for this. Chloride is classically considered one of the more resistant ion to meltwater-related processes. The larger ionic structure of Na+ is often cited as one of the main reason for which Na is highly incompatible for the ice molecular lattice, and is promptly excluded from solid ice. On the contrary chloride is more soluble in the ice and thus less affected by elution.

Reply

Citation: https://doi.org/10.5194/egusphere-2026-4779-RC2

Supplement

https://doi.org/10.5194/egusphere-2026-4779-supplement

Viewed

Total article views: 558 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
356	142	60	558	71	64	66

HTML: 356
PDF: 142
XML: 60
Total: 558
Supplement: 71
BibTeX: 64
EndNote: 66

Views and downloads (calculated since 23 Sep 2026)

Month	HTML	PDF	XML	Total
Sep 2026	162	40	19	221
Oct 2026	189	97	40	326
Nov 2026	5	5	1	11

Cumulative views and downloads (calculated since 23 Sep 2026)

Month	HTML	PDF	XML	Total
Sep 2026	162	40	19	221
Oct 2026	189	97	40	326
Nov 2026	5	5	1	11

Viewed (geographical distribution)

Total article views: 558 (including HTML, PDF, and XML) Thereof 558 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 02 Nov 2026

Short summary

Glaciers preserve records of past climate, but melting is rapidly damaging this information. We examined a new ice core from the Quelccaya Ice Cap in Peru by comparing a variety of chemical markers and using statistical tools. We found that melting has altered the archive mainly where water accumulated and near a fracture, rather than throughout the core. Much of the ice still preserves valuable climate information. Our approach can help assess other threatened glacier records around the world.


Total:	0
HTML:	0
PDF:	0
XML:	0