| 05 Jun 2026
Pre- and post-depositional processes affecting isotopic composition of seasonal alpine snowpacks – Austrian Alps
Abstract. Studying the formation and evolution of the seasonal Alpine snow pack is essential to obtain a clear picture of hydrological catchment processes during the snow season and the following melt period. In this study, three seasonal snow packs in one cirque in the Austrian alps were monitored on a weekly interval, the snow packs were allocated in line with each other at three different elevations (1500, 1600 and 1700 m a.s.l.). We analysed the establishment of isotopic signals using atmospheric observations and a back-trajectory model. In the post-depositional analysis, we tracked individual layers over time using Dynamic Time Warping to quantify the isotopic changes and relate these changes to the layer specific temperatures, temperature gradients and isotopic gradient individually, and also a multi-linear regression approach with the combination of temperature gradient and isotopic gradients. We also tried to relate the effects of global and net radiation on the upper snow layer isotopic metamorphism. We identified that meteoric first-order isotope signals correlate strongly with cloud conditions during snow accumulation events. The second-order stable water isotope interpreter, Deuterium excess (dxs), correlates strongest to moisture origin conditions. In the post-deposition analysis we concluded that during the stable period in the snow season trends in isotopic change are characterized by erratic week-to-week variations which always recover to the decreasing layer average. The magnitude of isotopic change increased with normalized profile depth, where the biggest fluctuations occurred at the snow-atmosphere boundary. In our analysis, week-to-week changes of the first isotopes cannot be explained by the snow temperature, nor the temperature gradient. The temperature gradient significantly affects layer dxs changes, driven by the preferential mobilization of δ18O rather than δ2H, contrary to classical sublimation-based expectations. The strong correlation between the isotopic gradient and isotopic layer metamorphism highlights diffusive homogenization as a dominant process, while temperature gradient induced vapour transport drives isotopologue differentiation. Upper layer snow isotopy is not affected by air temperature, but is affected by global radiation and net radiation. This study consolidates the idea that meteoric isotopy is shaped locally and non-locally for first- and second order isotopes, respectively. Furthermore, we highlight the coinciding processes of temperature gradient induced heterogenization of the snowpack and isotopic gradients induced homogenization.
Review for „ Pre- and post-depositional processes affecting isotopic composition of seasonal alpine snowpacks — Austrian Alps“ by Lammers et al. submitted to The Cryosphere
General comments:
This manuscript targets the understanding of the formation and evolution of the isotopic composition in seasonal snow. The authors present a temporal array of snow profiles from three different elevations taken during a 11 week window within one catchment in the Austrian Alps and with their analysis aim to decipher the drivers for isotope signal evolution. Their analysis confirms established theories for the formation of the initial isotopic composition of precipitation and characterize any “post-depositional modification” by comparing profiles taken on different days. They find that the most prominent post-depositional modifications are visible in the near-surface snow layer before the onset of melt. These signal alterations are correlated to environmental (radiation, air temperature) and in-snow variables (temperature-gradient, isotope-gradient) which form the basis for the identification of (directed and non-directed) diffusion as the driving mechanism for post-depositional processes.
Although the dataset presented is compelling and clearly the result of a substantial field effort, I think the manuscript would benefit from a clearer statement of motivation. Stable water isotopes are valuable tracers for hydrological processes, but it remains unclear what the authors intend to use these isotope signals for. Making this initial aim more explicit would help the presentation of the results and discussion connect more directly to a central research question, and would help bring out the manuscript's key messages.
The manuscript would benefit from research questions such as:
- What is the altitude dependent variability of the isotope signal and how does this affect the meltwater isotopic composition which can be used for hydrograph separation?
- Which isotope signals are preserved between precipitation and meltwater formation and how can we use this knowledge in hydrology?
- How does the isotope signal of snowpack and meltwater evolve from the onset of melt and how is this influencing hydrograph separation attempts using stable water isotopes?
In general I would recommend the authors concentrate on the melt-affected period of the dataset as this is a process not well established yet with in-situ observations.
In contrast, the temperature-gradient and isotope-gradient driven diffusion discussed are well-established physical processes known to influence isotopic profiles (Dietrich et al., 2023; Ebner et al., 2017; Gkinis et al., 2021; Merlivat, 1978; Touzeau et al., 2017); since they do not affect the bulk isotopic composition, however, I think their relevance to the manuscript's central aim could be motivated more clearly. In addition, a few statements regarding the “preferential mobilization of d18O rather than d2H” (Fig. 3) do not appear fully consistent with established principles of water isotope geochemistry (Merlivat, 1978). I believe this reflects how the results are being interpreted rather than an issue with the data themselves, and I detail this in the specific comments below, as I think resolving it would strengthen the manuscript's discussion of diffusive processes.
I would recommend that the authors add a clear, hydrology-relevant research question and revisit the interpretation of the diffusion-driven isotope signals discussed below. Addressing these two points would, I think, substantially strengthen the manuscript, and I would welcome the opportunity to review a revised version that incorporates them.
Specific comments:
- Sec. 4.1: The term “pre-deposition” indicates that water vapour, cloud particles or at a minimum very fresh snowfall was sampled and analysed; however, only snowpack samples were taken, which arguably are already “deposited”. If I interpret Fig 1 correctly, then snow samples that are analysed as “pre-deposition samples” are between 3- 6 days old and thus already affected by “post-depositional processes”. Therefore, I would argue that the authors cannot assess “pre-deposition” processes. The presented back-trajectory analysis and the correlation analysis with ambient variables from the reanalysis are a valid approach, but, as the authors point out in L.239, L.353, and Fig. S5, the surface (most recently deposited snow) is highly variable and sensitive to post-depositional processes acting on sub-daily and daily timescales and therefore does not preserve the original “pre-deposition” signal during 3-6 days. I would recommend to the authors to investigate how much of the original signal is preserved during their observation period, similar to (Aemisegger et al., 2022).
- Fig 1: The interpolated air temperature plotted here shows several strong and prolonged periods of positive degrees which would indicate melt-freeze processes affecting the snowpack in addition to pure dry snow metamorphism. Have the authors noticed melt formations in the snowpack, and if not, can we trust the interpolated temperature? I would be more interested to see the measured snow temperature rather than the interpolated air temperature.
- L. 110 - 114: Give more details on the installed thermistor string (model and manufacturer) how the installation was performed and how the sensors were calibrated. Show a comparison between the snow temperature measurements and the interpolated air temperature measurements. Since the sites are partially within a forest, local air temperatures might deviate significantly from the interpolated air temperatures.
- L. 279-280: Radiation is only one aspect of the surface energy balance yet the authors consider it as the driver for near-surface isotope variability. However, as radiation is not a phase-change inducing process the wording of this sentence is wrong. Research shows that instead of “solar radiation induced thermal gradients” it is the latent heat flux (i.e. sublimation) which leads to an enrichment of the snow surface (Beria et al., 2018; Hughes et al., 2021; Wahl et al., 2021; Ollivier et al., 2025; Stichler et al., 2001). A correlation with radiation is not surprising as it is part of the surface energy balance, but the causality between the two variables is not physical. The authors need to discuss the latent heat exchange at the surface in this context.
- L. 271-274 and Fig. 3, regarding the statement “preferential mobilization of d18O rather than d2H”: I think this description does not correctly capture the diffusion physics, and may reflect a misinterpretation of your own results (Table 3, Figure 3) rather than an error in the underlying data.
Under a temperature gradient, all three isotopologues (H2¹⁶O, H2¹⁸O and HD¹⁶O) diffuse simultaneously, each moving down its own vapor-pressure gradient (itself controlled by temperature) from the warmer towards the colder layer — none of them stays in place while another moves. Because δ18O and δ2H are both relative measures (ratios to the dominant isotopologue, H2¹⁶O), a shift in either delta value reflects a difference in diffusion rate relative to H2¹⁶O, not the “mobilization” of one isotope while the other remains stationary.
This has two consequences for the manuscript. First, in Fig. 3, depicting H2¹⁶O and H2¹⁸O as spatially separated by color is misleading: both isotopologues are present throughout the profile, and only their relative proportions shift. The main directed flux of H2¹⁶O should point consistently from the high vapor-pressure (warm) side toward the low vapor-pressure (cold) side, rather than "up and down" as the current schematic suggests. Second, because H2¹⁸O diffuses more slowly than H2¹⁶O, while HD¹⁶O diffuses only slightly more slowly than H2¹⁶O (the relative mass difference between H and D is smaller than between ¹⁶O and ¹⁸O), the two heavy isotopologues are retained to different degrees under the same temperature gradient. It is this differential retention — not a preferential mobilization of one isotope over the other — that produces the reduction in d-excess (kinetic fractionation). Please update Fig.3 to be consistent with the underlying physics.
I would suggest rephrasing the statement “dxs reduction under positive temperature gradients is driven by displacement of 18O towards colder layers, rather than by preferential 2H mobilization as conventionally expected” along these lines: both isotopologues diffuse under the temperature gradient, but at different relative rates, and it is this rate difference that changes the d-excess (see also Casado et al., 2021, for a similar framing).
- L. 244-248 “preferential mobilization of d2H relative to d18O into the liquid phase”: Similar to the previous comment, the authors here do not consider any sublimation or evaporation modifications in their interpretation of isotopic modifications after the melt onset. In addition to melt-freeze processes there will also be increased evaporation of liquid water and potentially still sublimation from snow which could explain the observed decrease of d-excess and increase of primary isotope values.
- Fig. 2: This is a very dense plot, and I think the information is not ideally conveyed. I would recommend staying with one snow-depth y-axis (should it be “height” or “depth”?) and add the isotope information as a colormap (add a colorscale) (see e.g.: Zuhr et al., 2023). Since the snow type is not used anywhere in the paper, I recommend removing this information here and adding the snow stratigraphy information in the appendix. I also recommend to indicate the melt or wet-snow period in this plot.
- L. 290: Discuss your results with the findings of (Aemisegger et al., 2022) who also sample isotope profiles repeatedly in seasonal alpine snow.
- Table. 5: I recommend to refrain from using a comparison with snow from the Antarctic plateau (Fujita and Abe (2006)). The water cycle processes and environmental conditions are very different, which leads to a highly non-linear d-excess behaviour (Dütsch et al., 2017)which are very likely not comparable to alpine seasonal snow d-excess-defining processes.
Technical corrections:
- Fig. 1: What is the measured snow temperature?
- L. 99: change “allocated” to “located” and delete “,respectively”
- L. 117: To me a “profile” sounds like a vertical profile through the entire snowpack height. However, the applied sampling strategy does not yield that. Please rephrase “profile” so that the horizontal sampling strategy is better described. The sampling cylinder used had a diameter of 8cm but Fig. 1 shows no continuous profile of 8cm resolution. How did the authors decide at which height / snowdepth to take a sample?
- L. 255: add a citation to that statement.
- Table 2 and 3: Please also show the correlation plots as the robustness of correlations are not easily to interpretable without a visual representation of the data. Are the R2 values given in Table 3 adjusted for multi-variable analysis?
- Change term “Rayleigh fractionation” to “Rayleigh distillation”
- L. 395: What is meant by the term “radiation-driven isotopic metamorphism”?
- L. 307: Change “suppressed” to “reduced”
- L. 308: Change “leading to a stronger relative enrichment of δ2H compared to δ18O” to “leading to a stronger fractionation of δ18O”
- Abstract, L. 6: “Dynamic Time Warping” is introduced without explanation or citation. Please briefly describe/cite the method here, or note that it is explained in the Methods section.
- Abstract, L. 8-9: The term “isotopic metamorphism” is used here without definition (see also comment on L. 395).
- Abstract, L. 16-17: This sentence restates the “preferential mobilization of d18O rather than d2H” interpretation discussed in the specific comment on L. 271-274; please revise it consistently once that interpretation is corrected.
Bibliography:
Aemisegger, F., Trachsel, J., Sadowski, Y., Eichler, A., Lehning, M., Avak, S., and Schneebeli, M.: Fingerprints of Frontal Passages and Post‐Depositional Effects in the Stable Water Isotope Signal of Seasonal Alpine Snow, JGR Atmospheres, 127, e2022JD037469, https://doi.org/10.1029/2022JD037469, 2022.
Beria, H., Larsen, J. R., Ceperley, N. C., Michelon, A., Vennemann, T., and Schaefli, B.: Understanding snow hydrological processes through the lens of stable water isotopes, WIREs Water, 5, https://doi.org/10.1002/wat2.1311, 2018.
Dütsch, M., Pfahl, S., and Sodemann, H.: The Impact of Nonequilibrium and Equilibrium Fractionation on Two Different Deuterium Excess Definitions, Journal of Geophysical Research: Atmospheres, 122, https://doi.org/10.1002/2017JD027085, 2017.
Hughes, A. G., Wahl, S., Jones, T. R., Zuhr, A., Hörhold, M., White, J. W. C., and Steen-Larsen, H. C.: The role of sublimation as a driver of climate signals in the water isotope content of surface snow: Laboratory and field experimental results, The Cryosphere, 15, 4949–4974, https://doi.org/10.5194/tc-15-4949-2021, 2021.
Ollivier, I., Steen-Larsen, H. C., Stenni, B., Arnaud, L., Casado, M., Cauquoin, A., Dreossi, G., Genthon, C., Minster, B., Picard, G., Werner, M., and Landais, A.: Surface processes and drivers of the snow water stable isotopic composition at Dome C, East Antarctica – a multi-dataset and modelling analysis, The Cryosphere, 19, 173–200, https://doi.org/10.5194/tc-19-173-2025, 2025.
Stichler, W., Schotterer, U., Fröhlich, K., Ginot, P., Kull, C., Gäggeler, H., and Pouyaud, B.: Influence of sublimation on stable isotope records recovered from high-altitude glaciers in the tropical Andes, Journal of Geophysical Research: Atmospheres, 106, 22613–22620, https://doi.org/10.1029/2001JD900179, 2001.
Wahl, S., Steen‐Larsen, H. C., Reuder, J., and Hörhold, M.: Quantifying the Stable Water Isotopologue Exchange Between the Snow Surface and Lower Atmosphere by Direct Flux Measurements, Journal of Geophysical Research: Atmospheres, 126, 1–24, https://doi.org/10.1029/2020JD034400, 2021.
Zuhr, A. M., Wahl, S., Steen‐Larsen, H. C., Hörhold, M., Meyer, H., and Laepple, T.: A Snapshot on the Buildup of the Stable Water Isotopic Signal in the Upper Snowpack at EastGRIP on the Greenland Ice Sheet, JGR Earth Surface, 128, https://doi.org/10.1029/2022JF006767, 2023.