Thermal Regime of High Arctic Tundra Ponds, Nanuit Itillinga (Polar Bear Pass), Nunavut, Canada

Young, Kathy L.; Brown, Laura C.

doi:https://doi.org/10.5194/hess-2023-297

Preprints

https://doi.org/10.5194/hess-2023-297

Preprints

28 Feb 2024

| 28 Feb 2024

Status: a revised version of this preprint is currently under review for the journal HESS.

Thermal Regime of High Arctic Tundra Ponds, Nanuit Itillinga (Polar Bear Pass), Nunavut, Canada

Kathy L. Young and Laura C. Brown

Abstract. This study evaluates the seasonal and inter-seasonal temperature regime of small tundra ponds ubiquitous to an extensive low-gradient wetland in the Canadian High Arctic. Pond temperatures can modify evaporation and ground thaw rates, losses of greenhouse gases and control the timing and emergence of insects and larvae critical for migratory bird feeding habits. We focus our study on thaw ponds with a range of hydrologic linkages and sizes across Nanuit Itillinga, formerly known as Polar Bear Pass (PBP), Bathurst Island, and whenever possible, compare their thermal signals to other Arctic ponds. Pond temperatures and water levels were evaluated using temperature water level loggers and verified by regular manual measurements. Other environmental data collected included microclimate, frost table depths and water conductivity. Our results show that there is much variability in pond thermal regimes over seasons, years, and space. Cumulative relative pond temperatures were similar across years, with ponds normally reaching 10–15 °C for short to longer periods except in 2013, a cold summer season when pond temperatures never exceeded 5 °C. Pond frost tables and water conductivities respond to variable substrate conditions and pond thermal patterns. This study contributes to the ongoing discussion of climate warming and its impact on Arctic landscapes.

Received: 17 Dec 2023 – Discussion started: 28 Feb 2024

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 preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 2492 KB)

Supplement (100 KB)

Download & links

Kathy L. Young and Laura C. Brown

Status: final response (author comments only)

RC1:
'Comment on hess-2023-297', Anonymous Referee #1, 19 Mar 2024

General Comments
This paper reports data collected over the course of nine years from Nanuit Itillinga (Polar Bear Pass) on Bathurst Island, Canadian Arctic. This is an “Arctic oasis”, a relatively productive wetland in the polar desert of central Canadian Arctic Islands. Compared to Greenland or Svalbard, the High Arctic of Canada is understudied, so this project provides a welcome contribution to our knowledge of wetlands in the Arctic. It is a relatively long-term study in an area where such data are difficult (and expensive) to obtain. The results of this study, especially data about year-to-year and pond-to-pond variability can be used in future modelling work on carbon budget, atmosphere-surface energy exchange, and so on.
No substantive issues with this paper. The approach is straightforward, the methods are the usual ones used for such studies and the data seem reliable. A couple comments on the writing and the figures below.
Specific comments
Line 12: drop "and whenever possible”
Lines 43-53: combine with previous paragraph?
Line 56: “…with focused pond studies in 2008 and 2009.” – reword?
Line 95: Drop “In this paper”, not needed; also, why not just say “We describe” rather than “we focus on describing…”
Line 97: “whereas" instead of “while”.
Line 99: “Substrate type varied across the ponds” is this within the pond, or the surrounding?
Line 112: “Less frequent manual measurements while manual estimates were made at distant ponds” Not clear
Line 122: “avoiding the flattening of data at low and high temperatures” not clear?
Line 125: “normality” or “normalcy”?
Line 126: “Given that autocorrelation did exist amongst the data (k-1) and is commonly found when comparing air to water temperatures (Johnson et al., 2014), no further work was carried out to develop a predictive model between air temperature and pond water.” Not clear this is an unsolvable issue. The autocorrelation of data is not an issue, it is the residuals in regression analysis. But also, can you remove the autocorrelation, perhaps using first differences or some other way? In any event, if you are not developing a model, then just drop this sentence?
Line 149: should this be in Discussion?
Line 153: Do you mean that Pond 1 and others show similar seasonal cycles? Reword?
Line 174: “Warming is comparable to that seen in Figure 2 …”, or something to that effect.
Line 178: Are these separate t-tests, and is there a multiple comparison problem?
Line 187: Mean July Temperature?
Line 188: Move to Discussion or Introduction.
Line 189-194. Drop sentence “Fig 5 plots…”. Move sentence from 194-195 to beginning of paragraph and add (Fig 5) at end. Since you mention here in the text the details of other data added, the caption within Figure 5 could be reduced to simply Dranga et al; Woo and Guan, Croft 2011.
Line 223-224 could be changed to: There is no significance difference in the cumulative relative frequency of periphery ponds across PBP (Fig 9).
Line 224: “The curve of South Small Pond…”
Line 234: “Figure 10 shows the …” Also, on the graphs, in 2008, most points are above the 1:1 line, except those near the origin; when pond temperatures are less than ~2oC, air temperatures are greater. Is this interesting, or simply noise? It is only a couple of points, but happens in all but East Medium Pond. It seems much rarer in 2009, but still sometimes seen.
Line 259: do you mean Pond 12 rather than 11?
Line 365: drop first sentence?
Line 288ff: Why not start with the next paragraph with your results, and then incorporate the results of this paragraph (lines 288-293) in relation to yours.
Line 307: Perhaps reword in the form: “The effects of warming and permafrost thaw on Arctic freshwater ecosystems remain poorly understood (Lougheed et al. 2011).”
Line 313: Reword to “The porosity of pond sediments depends on ice content” – is that what you mean?
Line 318: Is this sentence generally true, or only in relatively wet tundra systems? What about dry shrub tundra or polar desert?
Line 324: Perhaps explain more “… vertical pond seepage …”, Do you mean “…deeper thaw in warm or dry years contributes to more downward seepage of water into the deeper active layer which leads to drying of the pond.” – or change this for your meaning. And does it have to be both warm and dry or either?
Line 325: “... from one year to the next in the same pond …” And is it only colour or is it other aspects? Your supplemental figure shows grain size but is not mentioned in the text. It seems this is useful data, based on your conclusions, so perhaps move the figure into the text (unless HESS includes Appendix as part of the pdf of the paper, but at the end).
Lines 328-330: combine into one sentence? “In the Old Crow Flats, Yukon, Roy-Leveillee and Burn (2017) found that near-shore taliks could develop in shallow (often less than 20 cm) lake/pond water, when warm summers increased the thawing degree days.”
Lines 331-335: Not clear why you mention this. Needed?
Line 345: “conclusion” rather than “pattern”
Line 353: what do you mean by “terrestrial water inputs”? Do you mean “streamflow or groundwater rather than precipitation”?
Line 403: obtain rather than attain?
Line 453: Is the Lehnherr reference correct; the title and author list seem wrong?
Figures
Although acceptable, they can be improved to make it easier for the reader to study the results more quickly. However, I think it would be up to the author to decide if they want to make the effort, depending on how comfortable they are with graphics programming.
Fig 1: Lat-Lon lines may be useful on the upper-left map, at least. Also, the upper-left map is not labelled (a), (b) …; is this intentional? The a-b-c is confusing, and I am not convinced these letters are needed; the red squares seem to be enough.
In the graphs: Why are some lines dotted and others solid? Why are some thicker than others? It would be much nicer to make all solid lines, different colours and thinner. (The thick lines may be good for presentations, but in a paper, thin lines would be sufficient; too much is hidden behind the thick lines). It would also help if you maintained the same colour for the ponds, (when plotting them together), or the years (if that is what is being compared) between graphs. The journal can weigh in here, but these “pastel” colours don’t separate much, and bolder colours may be easier to see? But maybe this will create a problem with making accessible graphs. It would also be nice if the x-axis were the same for all plots – ie the same time span (ie 15 May - 15 Sept?), as this would allow for easier comparison between graphs.
Fig 2: Is the inset graph needed - doesn’t it just duplicate what is already on the graph? It would be nicer if the x-axis intersected the y-axis at -10, and maybe a horizontal, faint dotted line at 0oC if you wish
Fig 3: As above. Here is an example where you could use thinner solid coloured lines for the Pond temperatures, and use a dotted line for the Air temperature, to more easily indicate the difference.
Fig 4: It would help the reader if these were plotted the same as other - stacked on top of each other rather than 3 across. Vertically oriented x on the x axis is a pain to read.
Fig 5: Maybe make the points larger, although points would cover each other? But is there some other way to plot these? You can see year-to-year differences (cold vs warm) but distinguishing any one pond or seeing any relation to other properties of the pond (ie sediment colour) is difficult in the present form. And these symbols are a problem; for example Pond 6 could also be Pond 5 and 8 simply plotted on top of each other. Maybe a spaghetti plot would work? Or google to get some inspiration of other types of graph; many websites also include the necessary code (NOT bar graph, for example). What is CR1 and CR2? You may consider a light dotted horizontal line, or a light bar (including the s.e.) across the whole graph area at the appropriate value for the Dranga et al average, explaining it in the caption (if I am interpreting this correctly). What it the slash on the y-axis between 10 and 12? If it is from the Woo and Guan paper, then maybe the x-axis should go to 2004.
Fig 7: This stacked panels, with 2 columns is another way to present data from, for example, Figure 2, with each year in a separate panel. In that way, you avoid the issue of colour, although it is true that it makes it more difficult to compare between years. However, if the x- and y-axes are the same in all 7 graphs, and a faint grid is used, it may work.
Figure 8: Here you switch to Julian day. Is there a reason for the 2 different x-axes? This makes it difficult to compare between figures.
Fig 11: Why using points instead of lines? This makes it hard to distinguish between lines, as they cover each other. It is also “ugly”. On the y axis, it is not clear to which graph the -1.0 applies, so perhaps eliminate it, except on the bottom?
Fig 12: As in Fig 5 above. Maybe spaghetti plots may work? Or some other kind? On the vertical line separating the two panels, there are some lines. Are these from 2008 or 2009? Some separation would be helpful. And again, it seems that you should either use Julian or calendar dates for all graphs.
Data Files
Data are included as Excel files in a zip file. Within the zip, the files are labelled as “DRAFT”. Does this mean they are not final? If they are, then perhaps rename. Also, could you pass through the files and make sure all columns are well labelled, with units and if needed, with extra lines to explain clearly what each sheet and row/column includes. For example, in File “Fig10…”, sheet “2009 correlations” there are unlabelled columns. In the same File, sheet “2008 tair tw” there are unlabelled columns, and rows do not align. In File “Fig 9…” it is not clear what many of the columns are. This is a nice dataset, but it may be frustrating for people (or the authors in the future) to use it in the present form.

Citation: https://doi.org/10.5194/hess-2023-297-RC1
- AC3: 'Reply on RC1', Laura Brown, 04 May 2024
  
  The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2023-297/hess-2023-297-AC3-supplement.pdf
  
  Citation: https://doi.org/10.5194/hess-2023-297-AC3
RC2:
'Comment on hess-2023-297', Anonymous Referee #2, 25 Mar 2024

In this paper the authors summarize arctic pond temperature and chemistry data to highlight their sensitivity to warming. There are some nice data presented, and the analysis is robust. The paper could be improved with the addition of a conceptual figure near the end that synthesizes the data. Such a figure would make the paper more citable as it would provide a hypothesis for future work about how these systems respond to climate warming. Superficially, Sometimes the English was a bit clunky and a good proof read would fix this. My specific comments are:
Line 43: There is some off paragraph construction at this point in the paper.
Line 46: The authors juggle both Polar Bear Pass and Nanuit Itillinga at the beginning of the paper. I understand linking the two to avoid confusion with earlier scientific papers, but that only requires one instance. Perhaps pick one and go with that throughout. I suggest the Inuit name.
Table 1: I don't understand why "water table" is used. I assume this means "pond depth". I follow maybe where the authors are going with this, but maybe since "pond depth" is more appropriate in this situation.
Line 112: This sentence "Less frequent ...." is an example of the clunky writing.
Figure 3: Maybe lower the x-axis labels to improve clarity.
Line 176: "due to proximity" instead of "owing to nearness"
Line 180: "but it was different than West Medium Pond ...."
Figure 4: This figure would be more informative and better match the text if the panels were by year, each showing all three ponds.
Line 204: Could you re-do the frequency analysis with a common period available each year to address this problem?
Figure 9: Some simple statistics would help to objectively show these distributions are similar or different and this would help move the paper from a description of the data to an analysis of the data.
Line 344: Are not water levels dropping below ground and ponds drying out the same thing?
Line 353: Conversely, as rainfall increases, maybe there will be more runoff during summer and late summer. A few sentences and references about this would balance the discussion. Some references to lean on might be:
Beel, C. R., et al. "Emerging dominance of summer rainfall driving High Arctic terrestrial-aquatic connectivity." Nature Communications 12.1 (2021): 1448.
Bintanja, R., & Andry, O. (2017). Towards a rain-dominated Arctic. Nature Climate Change, 7(4), 263-267.
Line 369: Similarly, the discussion in this paragraph needs more attention than it gets. This is where the conceptual figure could come in. One that shows what warming and wetting (with a change in ppt phase) might do to these ponds.
Line 383: Maybe rephrase to "Increasingly warm and dry conditions may drain ....."

Citation: https://doi.org/10.5194/hess-2023-297-RC2
- AC1: 'Reply on RC2', Laura Brown, 04 May 2024
  
  The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2023-297/hess-2023-297-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/hess-2023-297-AC1
RC3:
'Comment on hess-2023-297', Anonymous Referee #3, 08 Apr 2024

General Comments:
In this paper the authors present a case study of the thermal regime of ponds in the Canadian High Arctic. Their study site is Nanuit Itillinga, formerly Polar Bear Pass, Nanuvut, Canada, and the data spans almost a decade (2007 – 2015). The data presented includes seasonal data of pond temperatures, cumulative relative frequency of pond temperatures, pond water specific conductivity, and frost table depths. This study is an important addition to expanding our knowledge of year-to-year variability of pond temperatures in the Arctic. Studies like these are rare, and crucial to document to expand our knowledge about Arctic tundra ponds and the climate change impact on Arctic landscapes. I don’t see any major problems with this paper. Below are some suggestions for improvement.
Specific Comments:
Line 54-63: I would like a couple of more sentences added to this section from information that is in Table 1. For example, “We studied X number of ponds, that ranged a surface area spanning from X-X”.
Line 180: add “Medium” after “East”.
Line 193: The Croft reference in the reference list is 2011, not 2013 as stated in the text.
Line 204-205: Wouldn’t you want to compare this for a similar time period? I am not sure you can compare between years if you are not using the same time period. I might be wrong on that, so ignore this comment if that is the case.
Line 234-239: Any thoughts on why r values are lower in 2009 compared to 2008?
Line 261: Any reason why the data isn’t shown in paper? I guess you don’t have to show it, but it would be nice to see that year’s data too.
Line 282: What is the temperature difference between the “cool” and “warm” season? Any idea on why there is such a big difference in specific conductivity? Is it tied to the temperature difference?
Tables
Table 1: This is a lot of information, and not much of this is mentioned in the text. It is a bit hard to envision how this will show up in print, but this table is currently three pages. Would this be better placed in the Supplemental? I am fine either way, so I will leave this to the editor. There is a parenthesis missing for the bulk density unit in year 2009 (Line 80-81). There are dates listed in some of the fields (Line 86-87) and not the others. Perhaps better to stay consistent throughout the table?
Figures
Figure 1: It is a bit confusing with the naming “a, b, and c”. There is no letter assigned to the top left figure. One option is changing top left to “a”: “Location of the PBP catchment on Bathurst Island, Nunavut (a) with the red outlined area zoomed in and shown in (b).” and so forth. Can you also add a scale bar to some of these figures? I understand this might be difficult for c (seeing as it is a picture), but it should be possible for the other images. Could lat and long be added to at least one of these maps? Is it possible to include the pond locations and numbers (e.g., Pond 1, Pond 2, and so forth)? This would give a visual on where these ponds are located.
Figure 2: The different thickness of the lines makes this figure (and other figures in the manuscript) a bit difficult to read. Is the inset needed? Also, why is the y-axis in bold letters? This comment is for all figures.
Figure 3: There is an overlap of 01-Jun with y axis. Remove? Add parenthesis for °C.
Figure 4: The discussion in the text is about comparing the data for location and each year. Would it be better to split these figures into years (2007, 2008, and 2009) rather than location?
Figure 5: This figure is a bit difficult to decipher with the different symbols (too small?), and some seem to overlap. Is there any other way this can be displayed? Maybe it is better presented in a table? In the legend, the Dranga et al. reference states it is from 1979-2009. I am only seeing one symbol in the graph for 2009. This might be because of an overlap of symbols. In the legend it says that the Croft, 2011 is unpublished. The reference shows this as a MSc thesis. So published?
Figure 8: Can you please add the date the picture (a) was taken? Also, you can add that to the graph in b. You have used dates in prior figures, should you use dates here as well to stay consistent? This same comment applies to the other figures using Jday.
Figure 11: The y-axis are overlapping between a, b, and c (you can’t see 0 in b and c). I suggest that you add spacing between figures. Do you have to have a negative sign for y-axis? I suggest removing the “-“. Also, why are there symbols? It could be a lot cleaner if lines were used instead.
Figure 12: Similar comments as Figure 5. The symbols are small, and it is difficult to read this figure. Add space between figures because of overlapping symbols.
Figure 13: remove units in legend. Units are in the y-axis. Instead of “Temp” and “SpCond” in legend, write “Temperature” and “Specific Conductivity”. There is plenty of room to fit that in the legend.

Citation: https://doi.org/10.5194/hess-2023-297-RC3
- AC2: 'Reply on RC3', Laura Brown, 04 May 2024
  
  The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2023-297/hess-2023-297-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/hess-2023-297-AC2
RC4:
'Comment on hess-2023-297', Anonymous Referee #4, 15 Apr 2024
This study conducts a detailed investigation of the thermal regimes of small tundra ponds in the Canadian High Arctic. Researchers collected important and valuable data, including sizes and hydrologic connections of ponds, to understand their variability. The research findings highlight the complexity of pond temperature dynamics and their broader implications for ground thaw, greenhouse gas emissions, and wildlife patterns. The study's focus on a specific region may limit its broader application, year-specific weather anomalies could skew long-term trend analysis, and uncertainties remain about the contributions of various water sources to the ponds.
Overall, this is an important work that contributes to our knowledge of the Arctic ponds. I will suggest it for publication after the authors address my comments and suggestions listed below.
I suggest adding a discussion on how the results from this study could be extrapolated to other regions. How is this study region geologically equivalent to other similar areas that have open water bodies?

Clark et al. (2020) applied the LAKE model to study the thermal regimes at three Alaskan lakes in a continuous permafrost zone. One of the major findings was that snow depth and lake ice period substantially influence water temperatures. It would be interesting to know the length of ice season for these lakes and how snow depth and early/late ice melt could affect the overall temperatures in the ponds, specifically for a cold summer season in 2013. If the ice period can be extracted from MODIS or related products, it would be interesting and extremely valuable to combine that data with the present lake temperature data.

In addition, I suggest including a description of future measurement activities at the ponds, if any. Will other data (e.g. GHG fluxes) be measured at these ponds besides thermal data?

Clark et al., Thermal modeling of three lakes within the continuous permafrost zone in Alaska using the LAKE 2.0 model, Geosci. Model Dev., 15, 7421–7448, https://doi.org/10.5194/gmd-15-7421-2022, 2022
Citation: https://doi.org/10.5194/hess-2023-297-RC4
- AC4: 'Reply on RC4', Laura Brown, 04 May 2024
  
  The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2023-297/hess-2023-297-AC4-supplement.pdf
  
  Citation: https://doi.org/10.5194/hess-2023-297-AC4

Kathy L. Young and Laura C. Brown

Supplement

https://doi.org/10.5194/hess-2023-297-supplement

Data sets

Replication Data for: Thermal Regime of High Arctic Tundra Ponds, Nanuit Itillinga (Polar Bear Pass), Nunavut, Canada K. L. Young and L. C. Brown https://doi.org/10.5683/SP3/KGRQDO

Kathy L. Young and Laura C. Brown

Viewed

Total article views: 470 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
353	72	45	470	36	22	27

HTML: 353
PDF: 72
XML: 45
Total: 470
Supplement: 36
BibTeX: 22
EndNote: 27

Views and downloads (calculated since 28 Feb 2024)

Month	HTML	PDF	XML	Total
Feb 2024	59	14	2	75
Mar 2024	106	23	10	139
Apr 2024	54	8	11	73
May 2024	91	21	21	133
Jun 2024	43	6	1	50

Cumulative views and downloads (calculated since 28 Feb 2024)

Month	HTML	PDF	XML	Total
Feb 2024	59	14	2	75
Mar 2024	106	23	10	139
Apr 2024	54	8	11	73
May 2024	91	21	21	133
Jun 2024	43	6	1	50

Viewed (geographical distribution)

Total article views: 468 (including HTML, PDF, and XML) Thereof 468 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 29 Jun 2024

Short summary

This work details the temperature and related variables of several High Arctic ponds in the Nanuit Itillinga (Polar Bear Pass) National Wildlife Area through nine seasons. The ponds show much variability in their temperature patterns over time and space. Ponds normally reached 10–15 °C for parts of the summer except in 2013, a cold summer season when pond temperatures never exceeded 5 °C. This study contributes to the ongoing discussion of climate warming and its impact on Arctic landscapes.


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