the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Sep 2024
The ISLAS2020 field campaign: Studying the near-surface exchange process of stable water isotopes during the arctic wintertime
Abstract. The ISLAS2020 field campaign during February and March 2020 set out to obtain a unique dataset describing the Arctic water cycle using stable water isotope (SWI) observations. Our observation strategy focused on measuring evaporation, deposition, and precipitation, all of which are commonly sub-grid scale processes in numerical weather and climate models. Uncertain parameterizations for these processes can lead to compensating errors, which can go unnoticed; however, evaporation and precipitation can also be investigated with SWIs, as they are an integrated tracer for processes that atmospheric moisture has undergone. The campaign can be divided into two efforts: the primary field experiment in Ny-Ålesund focused on evaporation and deposition, and the larger precipitation collection network around the Nordic Seas.
The primary field experiment lasted three weeks, from 23 February to 15 March 2020, with temperatures reaching below −30 °C. During these weeks, we obtained near-surface, high-resolution (approx. 20 cm) SWI profiles at two deployment sites. Using a newly developed profiling system, we measured SWI gradients in the lowermost 5 and 2 m over fjord water and snow-covered tundra, respectively. These profiles are complemented by fiber-optic distributed sensing (FODS) columns and nearby meteorological stations. The FODS columns supply continuous, high-resolution (2 cm or finer) temperature profiles above both locations, whereas the meteorological stations provide information on wind speed and direction. We also made a short deployment to the Zeppelin mountain observatory (472 m a.s.l) for measurements of the isotopic signal in the free-troposphere. Additionally, numerous water samples from the snowpack in and around Ny-Ålesund were taken, in addition to daily fjord water samples from Kongsfjorden. These samples provide the context for the surface conditions under which profiles were collected. Isotopic connections on the synoptic scale are achieved by linking Ny-Ålesund observations with precipitation sampling at locations across the European Arctic, namely Longyearbyen, Tromsø, Andenes, Ålesund, and Bergen. The resulting dataset provides comprehensive insight into the Arctic hydrological cycle and can facilitate the study of phase change processes and transport of water vapour into and out of the Svalbard region. Datasets from the field campaign are publicly available at the PANGAEA data repository (https://doi.pangaea.de/10.1594/PANGAEA.971241, Seidl et al., 2024).
- Preprint
(50021 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 13 Jan 2025)
-
RC1: 'Comment on essd-2024-293', Anonymous Referee #1, 29 Nov 2024
reply
Review of Seidl et al. “The ISLAS2020 field campaign: Studying the near-surface exchange process of stable water isotopes during the arctic wintertime” submitted to ESSD
This paper describes an extensive effort of water sample collection for isotope analysis in the Arctic during an exceptionally cold period in late winter 2020 with 11 near-surface profiles over a snow covered and an open ocean surface in Ny Ålesund as well as many precipitation samples from different location in Svalbard and along the coast on the Norwegian mainland. To me the near-surface profile data is truly innovative and interesting and the whole dataset provides exciting new insights into moisture cycling in the Arctic. I find the paper well written and very well documented. In some instances, the descriptions are a bit too detailed and lengthy, but in general I found the paper to be well-balanced. I would suggest providing an earlier overview figure of the sampled data together with the meteorological documentation at the beginning of the paper. Otherwise, the data and methods section is very challenging to read. But this can be easily solved by placing Fig.9 and 10 much earlier and relating them to Fig. 2 with the meteorological overview. I also recommend some more in-depth discussion of the isotope observations in the profiles to provide some physical consistency checks since the deuterium excess observations are indeed far from the normal range of expected measurements. Below you find some more detailed comments.
Minor comments:
- 6: after “primary” one expects to find a “secondary”, but was the precipitation sampling campaign really a secondary component of the field experiment? Or just at a larger scale?
- 12: meteorological station data?
- 36: put Pfahl et al. behind regional models and Brady et al behind earth system models.
- 39-40: “as well as the conservation of the isotopic imprint during further airmass transformation over open waters” what do you mean exactly by this? At different stages of the airmass transformation process? Or why “conservation”? Do you expect the isotopic imprint to be conserved during airmass transformation over open waters? I think this is a bit unclearly formulated. Also, to me it seems not so clear how you want to address this question. I really like the profiling aspect, but it’s definitely very local and I also like the distributed precipitation sampling, but to me it’s not so clear how you want to link them.
- 40: I would not write “validate” but evaluate. Otherwise, you start with a biased way of looking at the model assuming it does well in simulating the processes you are interested in.
- 44: d-xs is a strange notation, I would either opt for d or dexc, or dxs but the “-“ is confusing. After introducing it, the abbreviation should be used consistently (e.g. at L. 47 you write d-excess).
- 45-46: You use first non-equilibrium fractionation and then kinetic, that’s a bit confusing. Very likely the negative deuterium excess is also due to non-equilibrium fractionation but due to water vapour deposition in supersaturated conditions with respect to surface temperatures (Thurnherr and Aemisegger, 2022). You even write it yourself later at L. 54 (“freeze dried air masses”). This should be made clearer already here.
- L 53: Here maybe a reference to AC3 and Kirbus et al. 2024 ACP would be good.
Kirbus, B., Schirmacher, I., Klingebiel, M., Schäfer, M., Ehrlich, A., Slättberg, N., Lucke, J., Moser, M., Müller, H., and Wendisch, M.: Thermodynamic and cloud evolution in a cold-air outbreak during HALO-(AC)3: quasi-Lagrangian observations compared to the ERA5 and CARRA reanalyses, Atmos. Chem. Phys., 24, 3883–3904, https://doi.org/10.5194/acp-24-3883-2024, 2024.
- L45: “The HDO molecules are more likely to evaporate than the H218O molecules, a process known as kinetic isotope fractionation, which results in a positive d-excess signature” That’s not exactly true. HDO molecules are less likely to evaporate than H218O molecules. HDO has a lower saturation vapour pressure, and this effect dominates both the d18O and d2H signals. In equilibrium the d18O and d2H are related by a factor of about 1 to 8. It’s just that there is an excess of HDO in the vapour compared to the equilibrium case during subsaturated conditions.
- 42: I am not sure I can follow here: I don’t know of any theory that would have the ambition to predict the Arctic dexcess surface flux as such. Do you mean that we have an idea about the range of expected dexcess in the evaporation flux for a given range of temperatures and near-surface humidity gradients over the ocean? But there is still ongoing important controversy about the influence of snow metamorphism and what the isotope composition of the net snow sublimation flux is (see Wahl et al. 2024 TC).
- 42-50: Here I miss a statement that clearly summarises the literature already available on relevant processes. It has been shown in several recent studies (e.g. Thurnherr and Aemisegger, 2022, Brunello et al. 2024 GRL) that snow/ocean-air exchange is the key process that impacts dexcess in the mid- to high latitude marine boundary layer. The large-scale drivers of this dexcess variability has been linked to warm vs. cold air advection over the mid- to high latitude oceans. What is challenging in the Arctic and around Svalbard is that there are very large inhomogeneities in the surface conditions and both air-sea and air-snow interactions matter. In addition, not only airmass transport and temperature advection play a role such as within the core of the storm track regions, but also prolonged longwave cooling during stable anticyclonic conditions. During these conditions air-snow exchange can be enhanced due to snow metamorphism either within the surface snow (Casado et al. 2021 GRL) or during transport of blowing and drifting snow (Wahl et al. 2024, TC). So, I would say that there are already several literature building blocks available that provide the physical basis for establishing a more nuanced view for reading the Arctic dexcess signals. I suggest merging the paragraphs at L. 42 and L. 51 and discussing the large-scale drivers and physical processes directly with the associated known dexcess signals (near-surface vapour dexcess during upward vs. downward net fluxes, role of snow metamorphism).
- 59: Here the reader needs to know why this reconciliation between lab and field studies is necessary and what it entails. What do lab and field studies not agree upon?
- 76: the the
- Section 2.1: the deployment times don’t become clear from this section. How close in time where the three sites visited? Does the free tropospheric site really give a representative observation of the weather situation in which the observations at the two profiling sites were done? Also I didn’t find a figure showing these observations neither a discussion of how they could related to the observations at the profiling sites.
- Section 2.2: it would help to have the abbreviations of the sampling locations of Fig. 2 in the text as well. Furthermore, a timeline with an overview of the sampling periods for the samples taken at the different locations and sublocations would greatly help to get an overview of how much precipitation was sampled where and over which accumulation period. Figure 9 and 10 should be placed here and not in the results section.
- 149: A reference to the climatological work on Fram Strait CAOs and their preconditioning would be helpful here: Papritz et al. 2019
Papritz, L., Rouges, E., Aemisegger, F., & Wernli, H. (2019). On the thermodynamic preconditioning of Arctic air masses and the role of tropopause polar vortices for cold air outbreaks from Fram Strait. JGR, 124, 11033–11050. https://doi.org/10.1029/2019JD030570
- 155: the COAi has units of K or °C I assume.
- 158: the periods of profiling and where the profiles were done should also be listed clearly in Fig.2: I suggest splitting Fig. 2 into two: one figure covering the timeline in terms of sampling and meteo (as is) and another figure with a synoptic overview (where maybe 2 additional timesteps could be chosen, which are more representative for the profile sampling).
- 176: nearby the meteorological station
- Figure 3 and other locations: could the times be indicated in UTC or is there a reason for doing otherwise?
- 193-201: This is a very detailed paragraph on the FODS the data of which is published elsewhere, is this really necessary in the text? The deployment dates are relevant for the reader of this paper though.
- 210: the isotopic evolution of the snow
- Table 2: times of sampling are missing
- Table 5: what does this imply for the response time of the instrument?
- 350: “for a sample at -10 ‰d18O and -100 ‰ is thereby estimated as 0.44 ‰and 1.5 ‰, respectively” space between ‰d18O and missing dD indication.
- 374: at this time resolution the response time of the whole system is a key missing information.
- 408: double 10 s information
- Section 5.1 I think the profiles are exciting and THE big innovation of this paper: I would find it very valuable to provide the standard profiling plots for all the sampled profiles at the two sites in the supplement. Furthermore, the specific humidity is missing in the profiles, as well as the relative humidity with respect to the surface temperature (key variable to understand air-surface water isotope fluxes) and I would also be curious to see the d18O.
- Section 5.1: here the information on the snow and ocean isotope composition would be very important to have together with the profiles. That’s why they are useful, namely in combination with the surface profile observations.
- 450-460: I think for quality check reasons, the low deuterium excess data in air found here should be discussed in terms of their physical plausibility. Clearly the temperature profiles shown for the snow site indicate very strong heat and moisture deposition fluxes to the surface. There is some literature available on this kind of phenomena and their impact on the isotope signature of water vapour in polar regions:
- Negative values of the deuterium excess were also found in other studies e.g. during Mosaic and ACE and were associated with warm advection:
Brunello, C.F., Gebhardt, F., Rinke, A., Dütsch, M., Bucci, S., Meyer, H., et al. (2024). Moisture transformation in warm air intrusions into the Arctic: Process attribution with stable water isotopes. Geophysical Research Letters, 51, e2024GL111013. https://doi.org/10.1029/2024GL111013
And in the Southern Ocean:
Thurnherr, I. and Aemisegger, F.: Disentangling the impact of air–sea interaction and boundary layer cloud formation on stable water isotope signals in the warm sector of a Southern Ocean cyclone, Atmos. Chem. Phys., 22, 10353–10373, https://doi.org/10.5194/acp-22-10353-2022, 2022.
- In their Fig. 2 Thurnherr and Aemisegger, 2022 illustrate and explain the large amplitude change in dvapour compared to d18Ovapour during the process of water vapour deposition to the surface, although over the ocean.
- Other studies have discussed negative deuterium excess signals in polar regions as potentially due to sublimation:
Hu, J., Yan, Y., Yeung, L. Y., & Dee, S. G. (2022). Sublimation origin of negative deuterium excess observed in snow and ice samples from McMurdo Dry Valleys and Allan Hills Blue Ice Areas, East Antarctica. Journal of Geophysical Research: Atmospheres, 127, e2021JD035950.
- Wahl et al. 2024 hint towards the fact that fractionation due to air-snow interactions is likely not due to the sublimation part of the flux but to the depositional part of snow metamorphism (ongoing transformation of the physical structure of the snow important in particular in environments with a strong vertical temperature gradient), and show some evidence for the fact that the dexcess is lowered due to this process during a controlled wind tunnel blowing snow experiment.
Interestingly this profile was sampled during a CAO period over Fram strait. But the profiles with stable stratification for most of the time show that locally the site is influenced by a mesoscale wind system apparently advecting relatively warmer subsiding air over the snow site. I think these aspects should be highlighted because they matter for the credibility of the observations presented, which do deviate somewhat from the normal observational range for the dexcess.
- 462-476: Very nice illustration of a situation in which the closure assumption (i.e. that the water vapour isotope signal is the same as the isotope signal of the flux) is far from being satisfied. This could be mentioned here.
- Section 5.1: It is interesting to note that slight variations in the wind direction lead to substantial changes in the vertical temperature structure going from moist plumes over the ocean and subsiding air pockets likely leading to enhanced vapour deposition to more well mixed conditions over the whole column. Therefore, from what I see the temporal variability at one location is at least as large as the vertical variability sampled with the profiling arm. So individual eddies really dominate the temporal variations and the measured isotope signals at different levels. The profiling system is not fast enough to give insights into the vertical structure of one single dynamical feature of vertical transport. This aspect ought to be actively mentioned and discussed.
- Section 5.1: what does a wind direction change from 180° to 250° at the snow site imply for the air mass origin. How comes the vertical column is so differently stratified during these wind direction changes? Is there wind shielding or turbulence induced by the local infrastructure or by people?
- 6-8: error bars would be very helpful.
- Section 5.2: given the results from the previous section surveying conditions at the moisture source and then presenting the precipitation isotopes: I think a short discussion on the importance of the transformation of the signal underway due to fractionation, in particular, related to cloud processing would be helpful to tie the paper together and provide a more coherent storyline to the reader.
- 530: post-depositional modification of the isotope signals has been discussed to be due to snow metamorphism and quantified in several recent studies (e.g. Casado et al. 2021, Aemisegger et al. 2022, Wahl et al. 2024)
Casado, M., Landais, A., Picard, G., Arnaud, L., Dreossi, G., Stenni, B., & Prié, F. (2021). Water isotopic signature of surface snow metamorphism in Antarctica. Geophysical Research Letters, 48, e2021GL093382. https://doi.org/10.1029/2021GL093382
Aemisegger, F., Trachsel, J., Sadowski, Y., Eichler, A., Lehning, M., Avak, S., & Schneebeli, M. (2022). Fingerprints of frontal passages and post-depositional effects in the stable water isotope signal of seasonal Alpine snow. Journal of Geophysical Research: Atmospheres, 127, e2022JD037469. https://doi.org/10.1029/2022JD037469
- Good idea, one February 2020 event is described in Brunello et al. 2024.
Citation: https://doi.org/10.5194/essd-2024-293-RC1
Data sets
Calibrated stable water isotope measurements from Svalbard and coastal Norway during February and March 2020 [dataset bundled publication] Andrew Walter Seidl et al. https://doi.pangaea.de/10.1594/PANGAEA.971241
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 228 | 51 | 109 | 388 | 7 | 9 |
- HTML: 228
- PDF: 51
- XML: 109
- Total: 388
- BibTeX: 7
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1