Review of the article

“The coupling between hydrology, the development of the active layer and the chemical signature of surface water in a periglacial catchment in West Greenland”

The manuscript of Rydberg and colleagues presents actual results in studies of processes in permafrost hydrology, groundwater, chemical signature, and isotopes. The greatest advantage of the manuscript is presenting results of own investigations of periglacial lake’s catchment in a melting period. The presented field measurements and analytical results provide some important aspects and testify to the authors' knowledge of the hydrochemistry and modeling.

The authors apply quite a number of the different methods that could be better structured in the paper. Interpretation of the results is not clear enough due to complexity of overview, different timescales, or/and visualization. This paper provides new insights into periglacial surface and under-ground hydrological processes of Western Greenland, however, in chapter "Discussion" there is no enough discussion of some important questions in Polar hydrology in larger scale since these issues are important for the whole Arctic and not exclusively for Greenland.

General conclusion. The manuscript is recommended for publication after major revision according to following comments and notes. I believe the manuscript to be an interesting and good contribution. The manuscript will be useful for international Polar community on hydrology and complementary sciences studies.

General comments.

In “Objectives” authors hypothesizing about influence of freeze-thaw processes and hydrology (including water-age) to chemical and isotopic signature of surface water and groundwater in the lake catchment in West Greenland. This question is not new in permafrost hydrology. That is why then hypothesizing you should be precise with own study novelty justification. Better focus on own field measurements. There is not much information available in the literature concerning detail measurements of active layer hydrodynamic and chemistry during melting period. This is can be the core to build the paper around and shift the focus on analysis of the obtained data rather than modeling.

The manuscript needs to be partly re-organized. Obtained data and modeling could be presented separately. “Method” paragraph should be better structured as well as “Results”. Sometimes authors combine the data obtained by different methods, which significantly complicates the perception of the results. In my mind paper will benefit from separating the results of measurements and model calculations, and later you can discuss altogether in the “Discussion” chapter.

Figure 1 needs significant revision as it lacks much of what is mentioned in the text. The surface temporary streams should be shown on the map. Isolines of topography could be added as well for easier water flow direction understanding. The legend on the figure itself and the figure caption should be unified: “automatic weather station (AWS)” or “weather station”? “Ground temp” in the legend and “GW-12” in the texts has the same meaning? Is the point of “ground temperature” measurements and points of “lysimeter” and “groundwater well’s location” the same? Where is “AL-transects” on the map and in the legend? The lake name “Two-Boat Lake” should be added on the part B (there is no label “B” on the figure as well) and it should not be “TBL” as it is on the part “A”.

Figures numbers need to follow the order as it is in the text. Figures 3 and 4 should be rearranged to match the text. 

What type of vegetation zone is in the TBL catchment? Short explanation can be added to the paragraph 2.1. The type of vegetation is very important for evapotranspiration calculation and modeling as well as DOC variation analyze.  

Explanation of lyzimeter measurements should be provided in the text. Methods of soil/ground water sampling should be added in the text with more detail explanation: number of samples, time and depth of sampling, methods and instruments of water sampling; what kind of filters were used; how the samples were preserved and transported etc.     

Paragraph 2.2. During field work, repeated walking to take measurements and samples can by itself significantly affect the depth of thawing and even the chemical composition of the water. Was it taken into account in any way, either during field work or later in the data analysis?

The depth of TDR sensor location should be added into the text. The number of Ground temperature (GT) points are not clear on figure 1: there are 2 rings on a sector A and 3 rings (blue and black) on a sector C. Are all rings GT and TDR loggers? Could you add the information on a figure and to the legend? There is information about four locations on sub-catchment for TDR sensors (line 294), but no one of it is in the legend of Figure 1. This needs to be adjusted.  

The purpose of paragraph 2.4 of the Methods section (ground ice) is unclear as it is not subsequently presented in either the Results or the Discussion.

The model MIKE SHE and its applicability for permafrost zone should be better explained. For example, Line 161-162: “…four layers in the underlying permafrost (1-200 m depth)…”. Is this a typo? Could it be -200 cm? If this is indeed the case (200 m), then modeling permafrost to such depths requires a more detailed explanation. In that case, how it is used for surface runoff and groundwater (top 100 cm) modeling? Are the differences between thawing of permafrost and melting of ground ice? Did authors (or Johansson) validated the model for permafrost condition and specific hydrophysical processes? In my opinion, it should be highlighted in the methods. Referring to Johansson et al. (2015b) is not really enough since it is quite important for this particular study.   

Estimation of groundwater age is not fully clear due to used model type (Paragraph 2.6). MIKE SHE does not verified for permafrost conditions as it is mentioned in the text. The model has been tested for the boreal zone, which may be the reason for an inaccurate estimate of the groundwater age for the permafrost zone. Stable water isotopes content could help in a groundwater age estimation. The stage of hydrological cycle and, as a consequence, groundwater age directly influences the isotopic composition. Was the obtained data on isotopic composition analyzed in this regard? The age of groundwater was simulated for period 2016-2019 (Paragraph 3.2), that includes May 2017 when samples for isotope analysis were collected. 

The number of collected samples for water stable isotopic composition, unfortunately, is not really sufficient for statistical analysis. In this regard, it would be even more important to conduct a comparative analysis with known data on other Arctic regions.

Figure 5 (B) shows the graph of isotopic composition with only one value (point on the graph) for the rain and only one for the snow. Nevertheless, precipitation samples were collected in 2011, 2014, 2017, and 2019 as it is noticed in Paragraph 3.5. Are all the samples having the same value? Is it a mean value, may be? What about ground ice sample (Line 316) – what was the sampling depth? Line 317: “…winter (snow) and summer (rain) values…”. Why the rain strictly attributed to summer? Liquid precipitation occurring in spring (as in May 2017) and autumn and may have different isotopic composition.

Since DOC discussed in the text and provide an important information about water geochemical composition, more detail explanation is required (number of samples and sampling approach, filtration and conservation).

The figure 5 (A) with PCA analyze is fully overloaded. Lack of direct access to data does not allow reader to evaluate the reliability of the results. I would recommend to add the table with geochemical (and DOC) data as supplementary material or you may think about to place it to the open database, for example, Pangaea as used before (Johansson, Lindborg, Petrone et al.).

The main issue of a “Discussion” is water-age in a top active layer. The 1-year age is not surprising if you are sampling in May when snowmelt water partly flows down the surface and partly infiltrates to a ground. Meltwater is almost the only source for the runoff formation in May. The 3-4-years age of water is more interesting but the modeling results could be supported by stable water isotope values. Unfortunately, it could not be considered in the paper, possible due to lack of data. Relation of DOC and water age as well as vegetation and soil types are also not a novel just by itself. It seems to me necessary to elaborate on the formulation of the main conclusions, focusing on the novelty of this particular study.

Special Comments:

1. The legend to Figure 1 includes “Vegetation types” but “Water”, “Grassland”, “Barren”, and “Wetland” are not exactly vegetation types. Several vegetation complexes/microlandscapes could be on a wetland. The legend is recommended to be restructured better.  
2. The period of intense measurements could be added on the figure 2 for understanding of source of data - field or modeled.
3. The x-axis should be added on figure 5 (C).
4. The y-axis on figure 5 labeled as “Fraction of runoff from ground ice or rain” but it is different in the figure caption - “fraction of melting ground ice and rain…” The label of the y-axis should be unified.    
5. The figure 5 (C) is not really informative. Would be enough to discuss it in the text? The increase of groundwater fraction content of 0.2 in 12 days fits within the limits of measurement error.

Review of “The coupling between hydrology, the development of the active layer and the chemical signature of surface water in a periglacial catchment in West Greenland” for The Cryosphere.

General comments:

This is a well-orchestrated study of hydroloigic sources and flows into a small lake in southwest Greenland. The article would be of interest to permafrost scientists, hydrologists, and geochemists. The study design appears well done and is presented in a clear fashion.

However, as written this paper needs a lot of work- major revisions. There to many typographical errors, strange wording, and odd punctuation. It is confusing at times and it is not fair to ask a Reviewer to make the types of edits needed to make this more readable. I also have some issues with specific misuse of terminology (like “active layer”) and the statements ascribing evaporation as a main source of increased lake water ionic strength. This paper can probably be fixed but it will require some concerted efforts, more reading of similar literature, and a reframing of the results and conclusions.

Comments keyed to the text:

17: surface water

What is meant by “old groundwater”? Perhaps just say “ground water and shallow subsurface flow” here in the first sentence. A few lines later you say it is 4 years old so it is not old.

22: stable isotopic

23: what are hydrological situations? Events?

28: similar to other

31: Rainfall typically has little in terms of major ionic load. Look at your own data… Soil interactions are the dominant source of ions in waters. Thousands of papers show this. Including some in the Arctic/boreal.

32: through a catchment

33: and bedrock

37: projection instead of prediction. We do not “predict” like a Tarot card reader. We project results or model outcomes.

39: What is “the atmospheric state”? Conditions?

53: There are many studies in the boreal showing the source of this DOC is surface dead or dying organic matter and vegetation. For example:

Cai Y, Guo L, Douglas TA. Temporal variations in organic carbon species and fluxes from the Chena River, Alaska. Limnology and Oceanography. 2008 Jul;53(4):1408-19.

But this may be more germane since the study area is perhaps actually “quasi-boreal/tundra” (?):

Holmes RM, McClelland JW, Raymond PA, Frazer BB, Peterson BJ, Stieglitz M. Lability of DOC transported by Alaskan rivers to the Arctic Ocean. Geophysical research letters. 2008 Feb;35(3).

They show an increase in DOC with spring melt and with summer rain events. Also much of the boreal is discontinuous permafrost where many river shave year long flows. It may be better to focus on watersheds similar to the Greenland one that do not have substantial (or any?) groundwater flow in winter/late winter.

59: polygonal

60: Similarly,

63: and riparian

64: permafrost is not thawing in early spring. Do you mean thaw of the seasonally frozen and thawed layer?

65: I would put overland flow first as it is far more runoff volume than evaporation

How about “over the course of the summer thaw season” instead of “in time”?

82: the active layer is relatively thin in some areas but not everywhere. Maybe introduce continuous permafrost/high Arctic/cold here to frame the study catchments? I think the study area is in continuous permafrost yet many of the boreal studies used in the Introduction are discontinuous permafrost related. Does the study area have year-round groundwater flows? If not then, again, the Introduction needs to present studies of similar areas. Perhaps high Arctic Canada or Toolik Lake, Alaska or some of the Russian catchment studies?

This one:

Ma Q, Jin H, Yu C, Bense VF. Dissolved organic carbon in permafrost regions: A review. Science China Earth Sciences. 2019 Feb;62(2):349-64.

91: Despite being located

92: ) and little spatiotemporal data being available due to the region’s

94: means that the processes

95: covary

120: and active layer… were collected

125-125: this does not describe what is in “A”

126: red outline

127: I do not see a “B” labeled on the map areas. Use capital letters in the caption to be consistent.

128: C is to the right? What are the blue and black circles in C?

This Figure caption makes little sense…

138: no comma needed before were

139-141: has this “error estimate” protocol been used by anyone else? If so, it would be good to cite it. Surface elevation heterogeneity is probably a larger factor than anywhere “rising up” between measurements. What is the value of this error measured?

145: provide a summary of the thermistor depths

148: water was released

156: that are active

158: was estimated

165: zone from (no comma)

179-180: reach downstream

201: dissolved organic carbon

204: simultaneously with monitoring

207: and the snowpack

208: stable water isotopic

211: were thawed

221: data was then

223: indicating

223: they were not or they were? Why were they removed if they were not contaminated?

224: a groundwater well

259: the vertical line?

Use capital letters in the caption.

Do you have precipitation information? Rain events would be particularly good to add to this.

Figure 4. This is a really great representation!

285: equivalent

286: or evaporated? Perhaps as meltwater from snow pooled/moved across the landscape some was lost to evaporation?

294: no comma needed after sub-catchment

307: monitoring period

309: here and throughout: what you are measuring is not actually the “active layer” it is the depth of seasonal thaw at a given point in time. The true active layer would be the thaw depth in late August/early September. Please go through EVERYWHERE that you talk about the active layer and clean up spots where it is not the actual active layer being measured.

311: I recommend saying “stable water isotope(s)”

Figure 5B. Your stream water samples seem to be exhibiting evaporation from the “meteoric water line” identified as the line linking the snow and rain. This would be the preferential increase in d18O relative to dD in the samples.

Look at the d-excess values. The lower d-excess values in lake water suggest evaporation compared to rain and snow likely occurring between melt and flowing downslope to the lake. But I think over all the use of evaporation to get the lake ionic concentrations to where they are is overblown. Interactions with mineral particles/weathered surfaces is likely a larger source of ions.

Evaporation is not surprising at all but it must be considered as both a source of lost liquid (and thus higher major ion concentrations) as well as a means of altering the “mixing” between snow and rain and their relationship to the stream water and the lysimeter/ground ice. There are many citations you could pull in to address this. I am not sure how it would affect your mixing/fraction of melt calculations would be affected by this but you cannot ignore it.

348-350: I am not sure how you are stating the higher ionic concentrations in the lake are from evaporation. Yes, some evaporation is probably occurring in the lake and between melt/tundra and lake but the majority of the surface water likely is picking up ions along the way. Also, the bottom of the lake has a talik (as you say) and clearly the waters in the lake can mix with those mineral soils. As well as wind-driven mixing of soils and weathered materials during the open water season.

Maybe provide a plot(s) of some of the major ion concentrations in your different sample types- not just the PCA. Rainfall and snow typically have little to no ionic load. Attributing the increases in the lake form evaporation is likely not warranted.

For example, in looking at your data: Let’s use Ca.

Lake mean values: 2723

Precipitation mean values: 15717 micrograms/L

So that is a 7X enrichment… Is that much evaporation occurring? Probably not.

You could use Cl as a conservative tracer and look at Cl:X ratios of your ions. What does Cl do between precipitation and lake concentrations and is this all from “evaporation”?

Also, the isotope results show some evaporation but this is of the stream water not the lake. Likely the lake has some evaporation, too. Perhaps show the lake waters on the d18O-dD plot to see if there is fractionation that suggests evaporation?

357: on May 31^st

380: control

385: fall within

“both” is not needed (three items follow it)

386: in runoff

408: 0-25 cm what? Soils? Soil column?

412: here and throughout: unless the journal states otherwise references are typically presented in chronologic order. This seems to be reverse last name alphabetical?

420-424: No. The waters interact with surface soils and vegetation and pick up ions. This is not an evaporative process. You can calculate this: take whatever concentration something is in rain or snow and see how many “X” you would have to evaporate it to get it to the concentrations in the lake. You are totally ignoring soils and vegetation along the flow to the lake and with residence time in the lake as potential sources of ions.

Providing the concentration information would be helpful for the reader in assessing this further. Maybe plots for the SI? I may be wrong but the PCA and isotope data alone do not provide what is needed to calculate this.

424-425: water age means deeper footpaths/sources and thus more time interacting with soils and mineral particles as sources of ions.

429-430: EXACTLY! So why are you claiming it is all from some sort of evaporation process? This is where I am now confused… you spent so much time saying it is all about this mythical evaporation and now say it might be from the deeper flow paths. I agree with the deeper flow paths source.

434-435: again, look at that Cai paper and others. Increased river DOC from two sources/time periods: spring melt when there is a lot of movement of organic matter that degraded during winter. Then during summer rain events when more water flushed through slightly deeper flow paths into more organic matter sources.

511: are these ice wedges or segregated ice? Any idea how old this ice is? If they are ice wedges they are predominantly comprised of snow melt. If they are segregated ice layers then they likely represent more of a mean annual precipitation. If it is old then regardless of their ice type/formation source waters the colder climate had lower d18O and dD values for those sources compared to today?

513: movement of meltwaters during the winter how? The surface is frozen (as you say earlier).

522: consistent with patterns

536: has also

Data table: lots of negative values, particularly for precipitation. What are the analytical errors of the different measurements? Perhaps provide “<x” for anything that is below detection.