Summertime evaporation over two lakes in the Schirmacher oasis, East Antarctica

Shevnina, Elena; Vihma, Timo; Potes, Miguel; Naakka, Tuomas

doi:https://doi.org/10.5194/egusphere-2025-1964

Preprints

https://doi.org/10.5194/egusphere-2025-1964

Preprints

21 Jul 2025

| 21 Jul 2025

Summertime evaporation over two lakes in the Schirmacher oasis, East Antarctica

Elena Shevnina, Timo Vihma, Miguel Potes, and Tuomas Naakka

Abstract. The study quantified uncertainties in the bulk-aerodynamic method and combination formulas being applied in estimations of summertime evaporation over two lakes in the Schirmacher oasis, East Antarctica. The evaporation over the lakes was measured by the eddy-covariance (EC) technique during the austral summers (December–January) in 2017–2018 and 2019–2020. These direct measurements showed that summertime evaporation over two lakes varied from 0.3 to 5.0 mm d^–1. Depending on the ice cover presence, the average evaporation varied from 1.6 ± 0.1 mm d^–1 in December to 3.0 ± 0.2 mm d^–1 in January–February. In summer, the lakes were warmer than the ambient air, and the largest day-to-day variations in evaporation were associated with variations in the wind speed. The EC measurements were used as a reference for evaluating the uncertainties of the indirect methods. The bulk aerodynamic method gave the most accurate estimates of evaporation over two lakes (of 6–8 %), and this method showed acceptable skill scores (by two selected criteria) in estimation of the daily evaporation during the lakes' ice breaking-up and open water periods. This method is recommended for hydrological (lake water balance) applications required for operational (short term) decision making. Most of the combination formulas underestimated the summertime evaporation by 27–73 %.

Received: 25 Apr 2025 – Discussion started: 21 Jul 2025

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, 3720 KB)

Supplement (460 KB)

Download & links

Elena Shevnina, Timo Vihma, Miguel Potes, and Tuomas Naakka

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-1964', Anonymous Referee #1, 30 Aug 2025

The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1964/egusphere-2025-1964-RC1-supplement.pdf

Citation: https://doi.org/10.5194/egusphere-2025-1964-RC1
- AC1: 'Reply on RC1', Elena Shevnina, 21 Oct 2025
  
  Thank you for kind and useful comments.
  The manuscript was revised following your Specific Comments (SC) and our responses (R) are marked in below each comment.
  1. Does the paper address relevant scientific questions within the scope of HESS? Yes, unequivocally. The paper tackles the critical challenge of quantifying a fundamental component of the water balance—evaporation—in a sensitive and data-scarce polar environment. This directly aligns with the HESS scope of "physical, chemical, and biological processes within the hydrological cycle" and its emphasis on "the interaction of hydrology with other earth system sciences." Understanding these processes in Antarctica is vital for predicting freshwater availability for research stations, assessing the stability of ice shelves influenced by supraglacial lakes, and modeling regional climate feedbacks.
  R1. Thank you!
  2. Does the paper present novel concepts, ideas, tools, or data? Yes, primarily through its novel data. The core novelty is the presentation of rare, direct eddy-covariance measurements of lake evaporation in coastal East Antarctica. This dataset is a significant contribution in itself. The development and validation of a wind dependent parameterization for the bulk transfer coefficient (C_E) specifically for Antarctic lakes is a novel and valuable methodological outcome. While the concepts (EC, bulk method) are established, their application and rigorous validation in this extreme environment provide novel insights.
  R2. Thank you!
  3. Are substantial conclusions reached? Yes. The conclusions are robust, significant, and well-supported by the data:
  • Direct evaporation rates are quantified (0.3 to 5.0 mm d⁻¹), showing clear dependence on ice cover and wind speed.
  • Most combination formulas (Penman, Odrova, etc.) are shown to have severe systematic biases, underestimating evaporation by 27-73%.
  • The bulk-aerodynamic method is confirmed to be highly accurate (6-8% bias) but only when using appropriate, site-specific transfer coefficients (e.g., from Arya (1988)), not generic ones.
  • Wind speed is identified as the primary driver of short-term evaporation variability, a finding that contrasts with studies in less windy environments (like the Tibetan Plateau in Wang et al. (2019)).
  Wang, B., Ma, Y., Ma, W., Su, B., & Dong, X. (2019). Evaluation of ten methods for estimating evaporation in a small high-elevation lake on the Tibetan Plateau. Theoretical and Applied Climatology, 136(3), 1033-1045.
  R2.1: We added the following text after line 303: “Our results show that the wind speed is the primary driver for the short-term variation of evaporation, and it contradicts with the results for the lakes in Tibetan Plateau (Wang et al., 2019) where weather is, however, less windy than in coastal Antarctica”, and on line 569: “Wang, B., Ma, Y., Ma, W., Su, B., Dong, X.: Evaluation of ten methods for estimating evaporation in a small high-elevation lake on the Tibetan Plateau. Theoretical and Applied Climatology, 136(3), 1033-1045, 2019.”
  • The authors did not comment on the role of solar radiation which is the main driver of evaporation and needs to be discussed, even they did not directly measure it.
  R3. We added the following text after line 360: “The solar radiation is in the beginning of the causal chain of factors controlling ice and snow melt, lake water temperature and evaporation. It is explicitly included in the energy balance method which is among the other indirect methods applied to estimate the evaporation (Finch and Calver, 2008). In this study, however, we focused on indirect methods where the solar radiation is implicitly included in calculations because we did not measure the solar radiation in our experiments. ”
  4. Are the scientific methods and assumptions valid and clearly outlined? Yes. The methods are state-of-the-art. The use of EC as a reference is the gold standard. The post-processing pipeline (spike removal, footprint filtering, gap-filling) is clearly described and follows established protocols. The assumptions (e.g., the applicability of Monin-Obukhov similarity theory, the representativeness of point measurements) are standard for such studies and are clearly addressed. The statistical analysis using RMSE and SSC is valid and appropriate.
  R4: Thank you!
  5. Are the results sufficient to support the interpretations and conclusions? Yes. The results are comprehensive and compelling. The data from two different lakes and two summer seasons provide a robust basis for analysis. The figures (timeseries, diurnal cycles, scatter plots) and tables (method comparison, skill scores) effectively present the evidence. The clear gradient of performance across the different methods strongly supports the conclusion that parameterization is key. The finding that wind speed correlates better with evaporation than the vapor pressure deficit is convincingly demonstrated.
  R5: Thank you!
  6. Is the description of experiments and calculations sufficiently complete to allow reproduction? Yes. The description of the instrumentation, sensor heights, data processing steps, and equations is excellent. The provision of code and data on Zenodo is a major strength that ensures full reproducibility and aligns with best practices in open science.
  R6: Thank you!
  7. Do the authors give proper credit and indicate their original contribution? Yes. The introduction and discussion thoroughly contextualize the work within existing literature on polar hydrology and evaporation methods. The authors clearly reference the original sources of the combination formulas they test. Their own original contribution—the unique EC dataset and the subsequent validation of methods—is clearly stated and forms the central pillar of the paper.
  • Note while Wang et al. (2019) focused on a different environment, a discussion acknowledging that their finding (mass transfer methods work well) aligns with conclusions from other extreme environments (like high-altitude lakes) could further strengthen the context.
  R7: We included the text after line 303and on line 363: “Our results show that the mass transfer methods work well enough to reproduce the evaporation over the lakes in the Schirmacher oasis, and this is aligns with outcomes from the studies focused on the evaporation over the high-altitude lakes of Tibetan Plateau (Wang et al., 2019).”
  8. Does the title clearly reflect the contents of the paper? Yes. The title is accurate, specific, and concise, correctly reflecting the location, subject, and process studied.
  R8: Thank you!
  9. Does the abstract provide a concise and complete summary? Yes. The abstract perfectly summarizes the objectives, methods, key results (including quantitative findings), and the main conclusion and recommendation.
  R9: Thank you!
  10. Is the overall presentation well structured and clear? Yes. The paper follows a standard and logical structure. The flow is easy to follow, and the argument is built progressively.
  R10: Thank you!
  11. Is the language fluent and precise? Yes. The language is clear, formal, and scientific. While there are a few minor grammatical quirks (e.g., "containerizing" on p1), they do not hinder understanding. The manuscript is well-written.
  R11: Thank you! We have tried our best to improve the language.
  12. Are mathematical formulae, symbols, abbreviations, and units correctly defined and used? Yes. Formulas are presented clearly. Symbols are defined upon first use (e.g., in the bulk formula on p6). Units are used consistently throughout (mm d⁻¹, ms⁻¹, etc.).
  • Note: In Table 1, the column "Sum" has units "mm p⁻¹". This should be clarified to "mm per [33-day] period" to avoid ambiguity.
  R12: We corrected the text accordingly.
  13. Should any parts of the paper be clarified, reduced, combined, or eliminated?
  • Clarify: The distinction between "SSC" and "SSg" in the text and Table 4 should be made consistent.
  R13: SSg was the typo, and corrected in the revised version.
  • Clarify: The discussion of spray evaporation (p19) references "Eqs 3, 4", but only Eqs. 2 and 3 are presented. This should be corrected.
  R13: We corrected the text.
  • One needs to guess the applicability of Eqs 2,3, the meaning of the coefficients and the height where the wind speed w2 is measured should be explicitly stated.
  R13: The text was corrected.
  • L146: the formula for σ should appear before ‘where’ in L145.
  R13: Corrected.
  14. Are the number and quality of references appropriate? Yes. The reference list is extensive, relevant, and includes key historical works, foundational methodological papers, and recent literature. It appropriately covers the fields of micrometeorology, Antarctic science, and hydrological methods.
  R14: Thank you!
  15. Is the amount and quality of supplementary material appropriate? Yes, and it is a significant strength. The availability of the raw code and data on Zenodo is exemplary and exceeds typical standards. It ensures full transparency and allows for exact reproduction of the analysis, which is crucial for a validation study like this.
  R15: Thank you!
  In the revised manuscript, we implemented the modifications following the comments of three reviewers. We also try our best to smooth the language of the overall narrative and prepare new supplements with the modified code used for plotting the figures. The attached supplement shows the modifications implemented in figures and text.
  Elena Shevnina from behalf of co-authors
  
  Citation: https://doi.org/10.5194/egusphere-2025-1964-AC1
RC2:
'Comment on egusphere-2025-1964', Anonymous Referee #2, 11 Sep 2025

Comments on Summertime evaporation over two lakes in the Schirmacher oasis, East Antarctica by Shevnina et al.
In this study, evaporation was measured by the eddy-covariance technique during the austral summer two lakes in the Schirmacher oasis, East Antarctica. Both lakes are very small (only several hundred meters long and wide). The break-up durations of both lakes are short (only 1-2 month). So they are quite different from lakes in other areas of the world. Meanwhile, eddy-covariance (EC) technique is still rarely used in Antarctica. Therefore, it is an interesting study and worthy to be published in HESS. However, a major revision is still needed before publication. Some comments are shown below:
1, Result section is still unclear for me and should be reorganized. I am still unclear whether the measuring time span the whole ice-free period of the two lakes. The author should also show the result of both lakes one by one, so the readers can be easy to follow.
2, Section 3.1, the authors should clarify whether it is the whole ice-free period or not for both lakes. If the EC system was only operated only part of the ice-free period, the authors should also clarify. For example, Figure 5 shows that the lake was still ice-free on Feb 25, 2020, but the evaporation result only covers the period from 7 December 2019 to 8 January 2020 (Line 183). The authors should give clarify this in the revision.
3, Line 101-103, lake water temperature should be addressed in the next paragraph. Please reorganize this section and the following.
4, Section 4.2, The EC system was operated at both lakes. Evaporation over Lake Glubokoe was addressed, but the result of Lake Zub is not mentioned, why?
5, Figure 1, it is difficult to read. I can not find where are year round (red) and seasonal (blue) settlements from the Figure.
6, Figure 4, what do the vertical dashed lines mean? The start and end of ice should be marked in the Figure
7, Figure 5, The middle picture is taken from a different site compared with the other two.

Citation: https://doi.org/10.5194/egusphere-2025-1964-RC2
- AC2: 'Reply on RC2', Elena Shevnina, 21 Oct 2025
  
  Thank you for your useful comments. We implemented them in the revised manuscript, and the modifications are given in the supplement). We further respond to the Specific Comments (SC):
  SC1: Result section is still unclear for me and should be reorganized. I am still unclear whether the measuring time span the whole ice-free period of the two lakes. The author should also show the result of both lakes one by one, so the readers can be easy to follow.
  RSC1: The eddy-covariance measurements were collected on two lakes in two different experiments in the years 2017 – 2018 on Lake Zub and in 2019 – 2020 on Lake Glubokoe. We measured evaporation during the ice-free period for the first lake (Zub), and it was planned that the measurements for the second lake (Glubokoe) will cover the whole summer (ice break-up and free periods). However, from mid-January 2020 our instrumentation was technically broken, and the measurements only covered the ice break-up period for Lake Zub. The results for the first experiment (Lake Zub) were published in Shevnina et al. (2022), however, they lack deriving site specific transfer coefficients for the bulk-aerodynamic method. This study fills the gap, and also targets (a) to estimate evaporation from the direct measurements collected in the second experiment (Lake Glubokoe), and (b) to test the empirical coefficients for the bulk-aerodynamic and combination formulas developed out of the measurements collected in the two experiments. We added new text after line 62: “Also, site-specific transfer coefficients for the bulk-aerodynamic method have not yet been suggested, but this study fills the gap.”
  SC2: Section 3.1, the authors should clarify whether it is the whole ice-free period or not for both lakes. If the EC system was only operated only part of the ice-free period, the authors should also clarify. For example, Figure 5 shows that the lake was still ice-free on Feb 25, 2020, but the evaporation result only covers the period from 7 December 2019 to 8 January 2020 (Line 183). The authors should give clarify this in the revision.
  RSC2: The second experiment was planned to cover the whole summer period (ice-break-up and ice-free stages on Lake Glubokoe), but the logging system of IRGASON was broken in the middle of the season without possibility of being repaired in the field. We rewrote the text on lines 99–106 as follows: “The air temperature, barometric pressure, wind speed/direction and water vapour concentration were measured on a tower equipped with the EC open-path system IRGASON by Campbell Scientific. These measurements were collected on two lakes (Fig. 2 a) in two different experiments covering 38 days in 2018 and 33 days in 2019 – 2020. In the first experiment on Lake Zub/Priyadarshini, the evaporation was measured during the period from 1 January to 8 February 2018 when the lake was free of ice. The second experiment on Lake Glubokoe covered the period from 7 December 2019 to 8 January 2020. The experiment was planned to be carried out over the duration of the austral summer, but our eddy-covariance instrumentation was damaged in mid-January. Hence, the actual measurements on Lake Glubokoe only represented its ice break-up period.”
  SC3: Line 101-103, lake water temperature should be addressed in the next paragraph. Please reorganize this section and the following.
  RSC3: The text about the lake water temperature was moved to the next paragraph.
  SC4: Section 4.2, The EC system was operated at both lakes. Evaporation over Lake Glubokoe was addressed, but the result of Lake Zub is not mentioned, why?
  RSC4: This manuscript does not include the results for Lake Zub because they have been published previously in Shevnina et al. (2022). This previous paper, however, did not include (a) derivation of site-specific bulk-transfer coefficients and (b) performance of the empirical formulas evaluated against independent data. The present manuscript addressed these issues using the measurements collected on both lakes. All the above is clarified in the revised manuscript.
  SC5: Figure 1, it is difficult to read. I can not find where are year round (red) and seasonal (blue) settlements from the Figure.
  RSC5: Figure 1 and its legend were modified as follows: “Fig. 1. Location of the Schirmacher oasis (SA) (a) and its infrastructure (b): year round (red dots) and seasonal (blue dots) settlements connected by roads (yellow lines, © Humanitarian OpenStreetMap Team (HOT), 2020. Distributed under the Open Data Commons Open Database License (ODbL) v1.0.). © Google Maps, 2019. The red boxes in (a) and (b) outline the area with the main infrastructure in SA, and the image (c) shows the melted road to the White Desert Camp (photo D. Emelyanov, December 2019).”
  We replaced the text in lines 73-76 with the following text: “The oasis shelters two scientific bases operated year round since the 1960s (red dots within the red box in Fig. 1 b), and two tourist camps occupied seasonally (blue dots, Fig. 1 b). The scientific bases are occupied by 25–30 overwintering personnel, and up to 50 personnel during summer seasons. Up to 200 people can visit the tourist camps during summer. Two ice runways support transportation of people and cargo. Fuels are mostly delivered by ships to coastal bases located on the ice shelf, and then transported by vehicles to the settlements. The settlements, ice runways and coastal bases are connected with year-round ice roads (yellow lines in Fig. 1 b). In summertime, the transportation along the ice roads suffers from melted lakes and temporal streams formed over the ice surface. Figure 1 d shows the melted ice road to the White Desert Camp in December 2019.”
  SC6: Figure 4, what do the vertical dashed lines mean? The start and end of ice should be marked in the Figure
  RSC6: We improved Figure 4 and its legend in lines 162 – 170 as follows: Figure 4. The daily minimum, average and maximum for the air temperature (red lines) and LSWT (blue lines) measured on the shore of Lake Zub/Priyadarshini. The greed dashed lines show the beginning of the ice free period; the red lines on (b) show the dates of the lake images given in Fig. 6 b-d.
  We added the following text on line 171: “During most of the days in both experiments, Lake Zub/Priyadarshinis and Lake Glubokoe were warmer than the ambient air. During the period from 30 December 2018 to 9 February 2019, the mean daily LSWT in Lake Zub/Priyadarshini was 3.9 ºC, which was 4.7 ºC higher than the mean air temperature (Fig. 4 a). The difference between the LSWT and air temperature varied from –0.5 ºC (2–3 January, 2018) to 10 ºC (25–26 January, 2018). From 7 December 2019 to 15 February 2020, the mean daily LSWT of Lake Glubokoe was 3.1 ºC, ranging from 0.6 to 5.3 ºC. During this period, the lake was 2.4 ºC warmer than the air on average (Fig. 4 b). The largest difference between the LWST and air temperature was observed on 25–26 December 2019, when also the relative humidity was the highest, exceeding 90 %.”
  SC7: Figure 5, The middle picture is taken from a different site compared with the other two.
  RSC7. We changed Figure 5 and its legend on lines 172–180 as follows: “Figure 6. The instrumentation installed on Lake Glubokoe (a), © Google Maps, 2019; (b, c and d) show the panoramic images of the lake location of the HOBO logger (red dot) and the EC system (yellow dot) on 3 January, 14 January and 25 February 2020 (photos by Dmitrii Emelyanov).” We added the on line 181: “On Lake Glubokoe, the ice break-up period lasted from 8-12 December until 12-15 February, and the ice-free period lasted for approximately two weeks. The lake-ice cover was documented in a series of digital images taken from the three positions marked as Camera 1, Camera 2 and Camera 3 in Fig. 6a. Fig. 6b and 6c show the examples of the lake images taken on 3 January 2020 and 25 February 2020 using Camera 3 in different stages of the ice cover: the ice break-up (b) and the ice free (c). Fig. 6d was taken 14 January 2020 from the position of Camera 2. The EC measurements were taken during those 33 days when the ice had melted from 30–35 % of the lake surface. The fraction of lake ice was evaluated by processing the digital images of the lake ice cover (taken every 5-10 days in the period of December 2019 – February 2020).”
  In the revised manuscript, we implemented the modifications following the comments of three reviewers.
  Elena Shevnina from behalf of co-authors
  
  Citation: https://doi.org/10.5194/egusphere-2025-1964-AC2
RC3:
'Comment on egusphere-2025-1964', Anonymous Referee #3, 14 Sep 2025
The manuscript investigated summertime evaporation over two Antarctic lakes and is worthy in polar and hydrometeorology science. The observation data are valuable and the research methods are reliable, but the organization of the manuscript is not very coherent in logical. The narrative is relatively redundant. In particular, the discussion and conclusion sections contain too many contents that are repetitive with the previous text. The manuscript needs substantial revision:
Listing the four combination formula simply may improve understanding of the precision of the used 6 methods.

In Section 3.1, it may be better to introduce all of the information one lake by one lake.

The origin of the wind-dependent transfer coefficient (Eqs. 2-3, Fig. 10) is unclear. This is a critical part of the analysis and must be explicitly stated: are these relationships developed from the Zub 2017-2018 data in this manuscript, or are they presented as established relationships from a previous study? This has major implications for interpreting the validation results in Table 3 and Fig. 11.

If the C_Ein equation (2) and (3) were obtained in the previous work (Shevnina et al., 2022), it may appear in Section 3.2 Methods. Table 2 may be moved to the section, too.

Is Figure 10 from the above mentioned previous work? If not, how to get it?

The performance of the wind-dependent method needs explanation. If it was derived from Zub data, why does it not perform best for Zub? Why does the best parameterization differ between the two lakes? This warrants discussion on the site-specificity of these coefficients.

Figures: Most of subfigures and legends are unclear. It is recommended that each subfigure of a figure be labeled as a, b, c, d, etc., instead of distinguishing subfigures by terms like "Figure 2 top" or "Figure 2 bottom". Additionally, each piece of information in the figures should be explained in the legends.

1: The format of the numbering is inconsistent with that of other figures (e.g., Figure 2). It is suggested to unify the figure numbering format, such as using "Figure 1" consistently with "Figure 2".

Section 4.1: The descriptions in the text do not correspond to the figures, leading to confusion. It is suggested that the authors carefully check and revise this part.

Figure 3: I only observed black and green colors in the upper subfigure. Where does "water vapour concentration (blue)" in the legend come from? Does "relative humidity" correspond to green or black? Are the data in the figure daily data or half-hourly data? I assume they are half-hourly data, which would contradict the description in L152 that the range of daily air temperature is -4.9 to 5.1 ℃.

L196-203: This section is analyzed based on Figure 7, yet there is no reference to Figure 7 at all. It is suggested that the authors add "(Fig. 7a)" and other corresponding references at appropriate places in the text.
Citation: https://doi.org/10.5194/egusphere-2025-1964-RC3

Elena Shevnina, Timo Vihma, Miguel Potes, and Tuomas Naakka

Supplement

https://doi.org/10.5194/egusphere-2025-1964-supplement

Elena Shevnina, Timo Vihma, Miguel Potes, and Tuomas Naakka

Viewed

Total article views: 717 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
660	40	17	717	22	9	11

HTML: 660
PDF: 40
XML: 17
Total: 717
Supplement: 22
BibTeX: 9
EndNote: 11

Views and downloads (calculated since 21 Jul 2025)

Month	HTML	PDF	XML	Total
Jul 2025	72	9	3	84
Aug 2025	134	10	3	147
Sep 2025	424	13	8	445
Oct 2025	30	8	3	41

Cumulative views and downloads (calculated since 21 Jul 2025)

Month	HTML	PDF	XML	Total
Jul 2025	72	9	3	84
Aug 2025	134	10	3	147
Sep 2025	424	13	8	445
Oct 2025	30	8	3	41

Viewed (geographical distribution)

Total article views: 732 (including HTML, PDF, and XML) Thereof 732 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 21 Oct 2025

Short summary

The study first estimated the summertime evaporation over lakes located in coastal Antarctica with direct (eddy-covariance) measurements collected during two austral summers (December–January) in 2017–2018 and 2019–2020. The lake evaporation was on average 1.6 mm d^-1 in the ice break-up period, and it doubled in the ice free period. The bulk aerodynamic method with a site-specific transfer coefficient of moisture well reproduced the observed day-to-day variations in evaporation over lakes.


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