Stratigraphic noise and post depositional effects in isotopic record of Antarctic firn: insights from a snow trench at Dome C

Ooms, Adrien; Casado, Mathieu; Wahl, Sonja; Laepple, Thomas; Zuhr, Alexandra; Akers, Pete; Picard, Ghislain; Arnaud, Laurent; Savarino, Joel; Rehfeld, Kira; Stenni, Barbara; Dreossi, Giuliano; Masiol, Mauro; Landais, Amaelle; Masson-Delmotte, Valérie

doi:10.5194/egusphere-2026-4801

Preprints

https://doi.org/10.5194/egusphere-2026-4801

Preprints

25 Aug 2026

| 25 Aug 2026

Status: this preprint is open for discussion and under review for Climate of the Past (CP).

Stratigraphic noise and post depositional effects in isotopic record of Antarctic firn: insights from a snow trench at Dome C

Adrien Ooms, Mathieu Casado, Sonja Wahl, Thomas Laepple, Alexandra Zuhr, Pete Akers, Ghislain Picard, Laurent Arnaud, Joel Savarino, Kira Rehfeld, Barbara Stenni, Giuliano Dreossi, Mauro Masiol, Amaelle Landais, and Valérie Masson-Delmotte

Abstract. Water isotopes are a key tool to reconstruct past climate variations from ice core records. At sites with low precipitation amounts, such as in Central Antarctica, redistribution by wind and post-depositional processes can alter the snow isotopic signal. This causes a shift in mean isotopic composition and adds stratigraphic noise, with a large spatial variability at the meter scale. In turn, this can modify the isotope-temperature relationship at inter-annual timescales. Here, we present a new dataset of water stable isotope measurements performed on 39 simultaneous 1.5 m deep snow pits dug along a 50 m trench in December 2019 at Dome C. It is combined to timeseries of observed isotopic composition in precipitation and surface-snow conducted over the last decade in order to study the isotope transfer function during snow archival processes. The trench isotopic records covering about 20 years depict no annual cycle and no correlation with local temperature, contrary to precipitation and surface snow time-series. Trench profiles show a very large spatial variability of the inter-annual signal archived in snow. The trench mean δ 18 O values show a a +7 ‰ offset (-6 ‰ for d-excess) compared to mean precipitation values, and +2 ‰ offset for δ 18 O (-1 ‰ for d-excess) compared to surface snow, indicating vapour exchanges acting beyond the surface snow layer. The power spectra of trench profiles imply a stronger smoothing at high frequencies than predicted by classical diffusion models. Individual trench profiles are dominated by stratigraphic noise, with a signal-to-noise ratio less than 0.35 across frequencies. This signal-to-noise ratio can be enhanced by up to 50 % by inter-profile alignment based on chemical tracers. The aligned core stack signal shows a multi-annual signal with some features not reproduced by a simple precipitation stacking model, hinting to a contribution of post-depositional effects in this common signal.

Received: 09 Aug 2026 – Discussion started: 25 Aug 2026

Competing interests: At least one of the (co-)authors is a member of the editorial board of Climate of the Past.

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, 1645 KB)

Supplement (168 KB)

Download & links

Status: open (extended)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-4801', Anonymous Referee #1, 12 Oct 2026 reply
General assessment
This manuscript presents a valuable dataset of 39 isotope profiles from a 50 m trench at Dome C, along with precipitation and surface-snow observations. Assembling and analyzing such a dataset required substantial effort and deserves recognition. The spatial replication and alignment using independent chemical tracers are particular strengths. The study documents substantial spatial variability and systematic differences between snow and firn, and it provides useful constraints on how isotope variability is preserved on annual-to-decadal timescales in a low-accumulation environment. The individual sections are generally well written, but sometimes it is difficult to follow what is being compared, what is assumed for these comparisons, and what each comparison is intended to test.
My main concern is not the use of the VFC itself, which is explicitly designed as an idealized temperature-driven record of layered precipitation. Rather, I am concerned with the causal interpretation of discrepancies between this idealized precipitation and the trench, particularly the multi-annual anomalies. The prescribed daily precipitation-isotope composition in the VFC is a linear function of ERA5 2-m temperature and therefore omits observed isotope variability at fixed local temperature. The observed mean offsets still provide independent evidence for post-depositional modification. However, the time-varying VFC–trench discrepancies cannot, by themselves, be attributed solely to post-depositional processes. My major comments are as follows.
Major comments
Equations (1) and (2) make both isotope inputs to the VFC linear functions of ERA5 temperature: substituting Eq. (1) into Eq. (2) gives d-excess = −0.366 T₂m − 3.04‰. The VFC input therefore contains no temperature-independent isotopic variability. In Section 3.4, the authors report that temperature explains only part of the observed precipitation time-series variability (r² = 0.56 for δ¹⁸O and 0.20 for d-excess in the filtered data). Agreement in the precipitation mean does not establish that the model captures the anomalies relevant to the trench. In particular, the enriched δ¹⁸O layers in the trench (2003, 2014 and 2018) and the low-d-excess layer around 2015 are not adequately reproduced by the VFC (Figs. 1 and 3).　

The failure of the VFC to reproduce their amplitude and precise timing does not by itself establish that post-depositional processes produced them: the temperature-derived isotope forcing may also miss anomalies in the actual precipitation input. Please test this interpretation using observed precipitation isotope values where the records overlap. Before assessing these possibilities, the Introduction should not state that the discrepancy "can only be explained (L71-72)" by post-depositional processes: uncertainty in the temperature-derived precipitation isotope input is an alternative that remains to be tested.

It would be worth noting that Ollivier et al. (2025, The Cryosphere, Sect. 4.1.2) acknowledged that their artificial precipitation-isotope inputs did not accurately represent the observations and wrote: “we do not quantify the effect of post-depositional processes by directly comparing the model results and the observations”. At first sight, this appears to be a more cautious interpretation of model–observation discrepancies than the one adopted here. Please explain why the VFC comparison supports the attribution made in this manuscript, and state the assumptions and limitations of that interpretation explicitly.

Related to Comment 1: Eq. (2) imposes a fixed linear anticorrelation between δ¹⁸O and d-excess on the precipitation isotope input to the VFC, so both inputs vary only with local temperature. How well does that relationship describe the observed variability, rather than just an average tendency? Even at the same local temperature, d-excess may vary with conditions at the moisture source; this variability is absent from the VFC input. The manuscript acknowledges this limitation, but does not carry it through to the causal interpretation of the VFC–trench discrepancies.

It is difficult to understand the individual profiles from Figure 1. Please show the original δ¹⁸O and d-excess profiles as line plots against measured depth in the Supplement.

Minor comments
Part of the apparent δ¹⁸O–d-excess anticorrelation may reflect the linear definition of d-excess and the curvature of the meteoric isotope relationship (Uemura et al., 2012). Please test whether the main d-excess comparisons and interpretations persist when logarithmic d-excess is adopted. A water-isotope model also shows that, at very cold Antarctic sites, logarithmic d-excess is less dominated by condensation temperature and remains more sensitive to moisture-source temperature than linear d-excess (Markle and Steig, 2022). However, logarithmic d-excess is not a complete solution. Changing the definition would provide a sensitivity test, but would not address the absence of moisture-source variability in the temperature-only VFC input.

2. Section 2.4.3 first states that the diffusion length is estimated from the power spectral density (PSD), whereas Eqs. (9)–(10) are written in terms of spectral amplitude. The equations are consistent if amplitude was fitted, but it is unclear whether the PSD was converted to amplitude before fitting. Because power is proportional to amplitude squared, the Gaussian attenuation coefficient differs. Please check the equations used to infer diffusion length. (I raise this point because dividing the reported value of 4.0 cm by “square root 2” gives a value close to the 2.8 cm predicted by the Johnsen model. This agreement is not seen for the other estimates, so it may simply be coincidental.)

3. Section 2.1: report analytical precision for δ¹⁸O and δD, and d-excess.

4. Figure A7: the legend states that the blue estimate is based on 13 evenly spaced profiles, whereas the caption states that it is based on the 26-profile trench stack.

5. Line 249: -50.5 permil (i.e., the minus sign is missing).

References
Markle, B. R., and Steig, E. J. (2022). Improving temperature reconstructions from ice-core water-isotope records. Climate of the Past, 18, 1321-1368. https://doi.org/10.5194/cp-18-1321-2022
Ollivier, I., et al. (2025). Surface processes and drivers of the snow water stable isotopic composition at Dome C, East Antarctica - a multi-dataset and modeling analysis. The Cryosphere, 19, 173-200. https://doi.org/10.5194/tc-19-173-2025
Uemura, R., Masson-Delmotte, V., Jouzel, J., Landais, A., Motoyama, H., and Stenni, B. (2012). Ranges of moisture-source temperature estimated from Antarctic ice cores stable isotope records over glacial-interglacial cycles. Climate of the Past, 8, 1109-1125. https://doi.org/10.5194/cp-8-1109-2012

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4801-RC1

Supplement

https://doi.org/10.5194/egusphere-2026-4801-supplement

Viewed

Total article views: 380 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
219	120	41	380	81	49	44

HTML: 219
PDF: 120
XML: 41
Total: 380
Supplement: 81
BibTeX: 49
EndNote: 44

Views and downloads (calculated since 25 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	41	23	4	68
Sep 2026	104	57	31	192
Oct 2026	74	40	6	120

Cumulative views and downloads (calculated since 25 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	41	23	4	68
Sep 2026	104	57	31	192
Oct 2026	74	40	6	120

Viewed (geographical distribution)

Total article views: 377 (including HTML, PDF, and XML) Thereof 377 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 26 Oct 2026

Short summary

This work presents new insights in the archival process of multi-annual climatic signal in the near surface firn at Concordia Station, East-Antarctica, from the new data set of high resolution stable water isotope profiles measured in a snow trench. The dated isotopic profiles covering 20 years are compared against overlapping precipitation and surface snow observations. This highlights the strong contribution of stratigraphic noise and post depositional effects to individual trench profiles.


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