the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Aug 2024
Diatom shifts and limnological changes in a Siberian boreal lake: impacts of climate warming and anthropogenic pollution
Abstract. Lake ecosystems are affected globally by climate warming and anthropogenic influences. However, impacts on boreal lake ecosystems in Siberia, remain largely underexplored. Our aim is to determine if shifts in diatom assemblages in a remote lake in eastern Siberia are related to climate warming, similar to observations in temperate regions, while also exploring how the ecosystem might be influenced by climate and pollution through various biogeochemical proxies. We analysed continuous sediment samples from a 210Pb–137Cs–dated short core from Lake Khamra (59.99° N, 112.98° E), covering 220 years (ca. 1790–2015 CE), with the specific feature of combining a variety of proxies on the same sample material to provide a comprehensive record of environmental changes in this less–examined region. Biogeochemical proxies include carbon and nitrogen concentrations (TOC, TN) and corresponding stable isotopes of bulk sediment samples (δ13C, δ15N), as well as diatom silicon isotopes (δ30Sidiatom), alongside light microscope diatom species analysis. The diatom assemblage at Lake Khamra is dominated by few planktonic species, primarily Aulacoseira. At 1970 CE, we observe a significant shift in diatom assemblages, with a marked increase in the planktonic species D. stelligera and a decrease in Aulacoseira, which we attribute to recent global warming, earlier ice–out, and potential enhanced summer thermal stratification, aligning with similar changes seen in temperate lake ecosystems. Furthermore, we see evidence for an increased diatom productivity supported by rising diatom valve concentrations and accumulation rates. Carbon and nitrogen levels increase in the 1950s, preceding the 1970 CE shift in diatom assemblages, suggesting that hydroclimatic and fire–related changes in the catchment significantly influence the limnology. Increased precipitation and weathering are discussed to alter silica sources leading to decreasing δ30Sidiatom after 1970 CE and suggest δ30Sidiatom as a proxy for weathering rather than productivity at Lake Khamra. Indications of human impact on the lake ecosystem include a 13C–depletion, linked to fossil fuel combustion since the 1950s, and changes in diatom species composition, such as the increased abundance of planktonic A. formosa. Furthermore, we observe a clear acidification trend since the 1990s, marked by a drastic increase in Mallomonas scales. Strong correlation to mercury accumulation rates, determined in a previous study, indicates a long–distance air pollution trend. We conclude the ecosystem of Lake Khamra is profoundly affected by climate warming and human–induced pollution, emphasising the urgent need for comprehensive research to address and mitigate these impacts on remote lake ecosystems to secure natural water resources.
- Preprint
(1690 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 02 Oct 2024)
-
RC1: 'Comment on egusphere-2024-2470', Anson Mackay, 12 Sep 2024
reply
Stieg review BioGeosciences
- Does the paper address relevant scientific questions within the scope of BG?
BG seeks interdisciplinary studies that look at the interactions between e.g. biological, chemical and physical processes. Steig et al. do this by taking a palaeolimnological approach to investigating biological and organic geochemical records in a relatively remote region of eastern Siberia.
- Does the paper present novel concepts, ideas, tools, or data?
High resolution diatom analyses and organic geochemical analyses are well established in palaeolimnology, including multiproxy studies. However, there is novelty in these data in the presentation ofd 30Si of the diatom silica in the Lake Khamra sediment record. Records of d 30Sidiatom are still relatively uncommon in lake sediment records, yet they have the potential provide much needed information on e.g. lake biogeochemistry, dissolved silica utilisation etc. What makes their record stand out even more, in my opinion, is that the data are presented at a decadal resolution for the past 200 years.
I think that there is also a missed potential of presenting these d 30Sidiatom records alongside their d18Odiatom records from a previous publication – having joint isotopes from the same diatom material, alongside the diatoms themselves, has great potential for quite deep lake-catchment ecosystem understanding.
Moreover, although d15N data are presented, they are barely discussed at all, which I found surprising given the discussion on potential acidification of the lake, changes in atmospheric reactive nitrogen etc.
Just thinking about it more, their samples actually have 4 different types of stable isotopes investigated but little is made of this.
- Are substantial conclusions reached?
Substantial conclusions are reached, but I don’t agree with one of the main findings that their record provides evidence for lake acidification.
- Are the scientific methods and assumptions valid and clearly outlined?
Largely yes, except for the use of chrysophyte scales to create an index to be interpreted as a record for lake acidification.
- Are the results sufficient to support the interpretations and conclusions?
Largely yes, but not for lake acidification
- Is the description of experiments and calculations sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)?
Yes
- Do the authors give proper credit to related work and clearly indicate their own new/original contribution?
Yes
- Does the title clearly reflect the contents of the paper?
I don’t think that their study really does focus on anthropogenic pollution. No pollutant data are actually shown (only referred to) and I question the validity of using a chrysophyte index as their evidence for lake acidification.
- Does the abstract provide a concise and complete summary?
It will need to be revised
- Is the overall presentation well structured and clear?
It makes sense to present data in order that these data are then discussed.
- Is the language fluent and precise?
Overall yes, but the authors needs to be more careful in use of e.g. ‘statistical’ when no statistical tests have been done, and ‘various’ which leaves the reader to make guesses. And sometimes statements are rather vague, or need further supporting evidence. I have highlighted these where relevant below and in the annotated PDF
- Are mathematical formulae, symbols, abbreviations, and units correctly defined and used?
I think I found one mistake in how the diatom fluxes are presented – details are given below.
- Should any parts of the paper (text, formulae, figures, tables) be clarified, reduced, combined, or eliminated?
I think that the figures need to be re-ordered so that they are presented in the order that they are subsequently discussed in. The discussion is overly long and could be more focussed
- Are the number and quality of references appropriate?
Largely yes, although I think that there are many more studies from Siberian lakes that have looked at recent environmental change. Also, I’d like to see better use made of studies that detail diatom autecologies -these will help in the diatom interpretations.
- Is the amount and quality of supplementary material appropriate?
If my suggestions for corrections are found useful, then I don’t think any SM is necessary
Diatom shifts and limnological changes in a Siberian boreal lake: impacts of climate warming and anthropogenic pollution
Overall review
This is an interesting palaeolimnology study that provides much needed information in lake ecosystem changes in a relatively remote region of eastern Siberia. The authors take a multiproxy approach to understand how environmental change has impacted Lake Khamra over the past 200 years. The datasets are excellent, with radiometric dating suggesting that a decadal resolution has been reached throughout the time-period investigated. This is no mean feat, as lakes in these cold regions often have very slow accumulation rates, making records of recent change difficult to decipher. This is not the case here.
The main datasets presented include biology (diatoms) and organic geochemistry. These are palaeolimnological mainstays. However, there is also an excellent dataset of diatom silicon isotope analyses d30Sidiatom, that is of decadal resolution. There is also a record of Mallomonas chrysophyte scales, but more on that below. The study details ecological and geochemical changes over the past 200 years that are concurrent with other studies in the region and further afield, even in different continents (e.g. North America, northern Europe) of the impact of warming on ice-covered lakes. A convincing key finding from Steig et al. is a shift in geochemical properties of the lake (ca. 1950 CE) that precede shifts in diatom communities a couple of decades later (ca. 1970 CE), trends that are exhibited in other cold regions of the northern hemisphere.
Less convincing however, is the suggested impact from pollution, causing the lake Khamra to acidify in recent decades – here I think the study is not critical enough in the evidence presented. Indeed, as detailed below, there a few parts of the study where a greater and more critical examination of the literature is required.
Overall, I recommend publication, but major revisions will need to be made. I provide more detailed comments below, and in the attached PDF.
Specifics
Title: As will be clear below, I’m not sure that the title is appropriate with regards to impacts of anthropogenic pollution. I didn’t see any keywords on the manuscript, but I think the term ‘multiproxy’ would be a good one to add, either to the title or as a keyword.
Introduction:
Line 49- : the statement “These lake observations align with the proposed onset of the 'Anthropocene', a geological epoch beginning in the mid–20th century, introduced by Crutzen and Stoermer (2000) and further supported by Zalasiewicz et al. (2017).” Is rather out of place, now that the Anthropocene has not been retified as a formal Epoch. Use of the term Anthropocene is still of course valid, but requires a more nuanced context.
Lines 56-57: I want to push back a bit on the statement “However, there is a notable gap in understanding how these global changes impacted remote and less studied regions like Siberia.”. I think that there are a considerable number of studies showing evidence of recent environmental change over the past 150 years in Siberia that could be used to here, but I do agree that much more research is needed for this region.
Lines 65-66: you attribute “reduced lake ice cover duration” and “stronger thermal
Stratification” since the 1950s to Roberts et al. 2018, but both those statements are linked to work done by other studies that Roberts et al. have cited. I think my point here is to make sure that when you cite a study, you cite it for the research it itself has done.
Line 75: Leng and Barker reviewed oxygen isotopes in diatom silica, not silicon isotopes.
Line 82: I don’t think I would call Hg a biogeochemical proxy
Line 90: “…the lakes investigated are either not located in Siberia…” but this is not really true. For example, even from my own (and collaborators) and other Russian-led work, there are many studies looking at palaeo records in Siberia for past 150 years.
In this section of the Introduction, in providing a setting for their study, the authors justify why their study needs to be done by saying where other studies have been done, with few in Siberia. But the case made is not convincing as sources used are rather old and not comprehensive. I think the authors should have confidence that their own study importantly adds to a growing body of work from Siberia and the Yakutian region especially. Maybe a route through this would be to focus in on the Yakutian region more, as Siberia is vast, over many eco-regions N-S, E-W etc, with many competing drivers of environmental change.
Lines 94-98: With the sentence starting “Beside oxygen isotopes, silicon isotopes of diatoms…” the rationale for Si can be better expressed as an indicator of changing productivity and nutrient utilization etc, which has been found in other lakes, including Baikal (see Virginia Panizzo-led articles).
Line 99: be careful using terms such as “significant” when significant tests were not undertaken. This is done many times throughout the manuscript
Lines 104-107: with the objectives, I’m not sure that erosion input linked to climate variability has been shown, as no proxy for erosion has been specifically measured (possibly % dry weight might work here, but something like magnetic susceptibility would have been a better indicator). Overall, a much more critical approach needs to be taken on the impact of pollution on diatoms in the lake.
Section 2: Overall, the methods section is excellent – really well written.
Line 159: state / summarise how water content and DBD were determined, (with full details given in Stieg et al. 2024a). For example were these determined through a scanner or through physical drying in an oven). I think it would be good to have MAR shown in one of the profiles.
Section 2.4: This really is an impressive highly-resolved dataset
Lines 222-223: the range of flora used are quite European focussed. Were any other guides used, especially Russian flora?
The only approach I have an issue with is the use of a Mallomonas index as a proxy for lake acidification. Greater justification for lake acidification needs to be made, because I was quite surprised to hear that acidification could occur give that the catchment consists of “Cambrian bedrock composed of alternating dolomite and limestone”, rocks that are rich in acid-neutralising cations. In the introduction to the study, while the threats of climate change are well set out, it is not made clear if atmospheric pollution and lake acidification are real threats to the region.
Section 2.6: lines 261-264, is there a mistake with the units?
If MAR = g/cm2/a, and concentrations are 10^7 valves/g, then that will mean DVAR = 10^7 valves /g/cm2/a. I assume the authors are multiplying by 100 to give 10^9 valves /g/cm2/a (although I’m not sure why!). But DVARs are then being expressed as 10^9 valves /m2/a and this is wrong - it should be 10^9 valves /cm2/a. The /m2 was calculated for reporting the changes in carbon fluxes. If you were to go from /cm2 to /m2 you'd need to multiply the diatoms by 10,000? Or am I missing something here?
Line 263+ 266: as the data being correlated are from a sediment core, they are majourly affected by temporal autocorrelation and not truly independent. Therefore, Pearson correlation is not appropriate here. Moreover, as mixed data types are being correlated (and data not tested for normality), a Spearman-Rank correlation would be more appropriate. Lastly, I'd suggest a more rigorous p value as your data are not independent e.g. 0.005 at least. I’d actually question of correlating these data to such an extent is even warranted as I don’t think this analyses adds to the study.
Lines 288-290: There is an interesting decline d15N since the 1960s -is this similar to studies of remote cold lakes in e.g. North America where a decline in also observed, and if so, is it related to increase in isotopically lower nitrogen from anthropogenic sources? (I note however that the magnitude of decline in Khamra is much lower than in other studies).
Figure 2: an effective way of comparing geochemical trends with the diatoms would be to plot here the PCA axis 1 and axis 2 sample scores. I see that these are presented in Figure 4.
Figure 3: Are the authors certain Fragilaria cf. gracilis is tychoplanktonic (where was this information obtained from)? Fragilaria cf. gracilis has recently been suggested to actually be F. radians. If that is truly the case, then this would mean F. gracilis/radians a planktonic taxa not tychoplanktonic. https://fottea.czechphycology.cz/artkey/fot-202202-0009_fragilaria_radians_k_tzing_d_m_williams_et_round_the_correct_name_for_f_gracilis_fragilariaceae_bacillari.php#:~:text=Fragilaria%20gracilis%20is%20one%20of,rivers%20to%20even%20eutrophic%20lakes.
Overall a succinct account is given of the main changes in the diatom zones.
Figure 5: This is a nice figure. But I think the colours for the triangles for the two time periods could be more clearly contrasted (even with different shapes)
Discussion
The Discussion is overly long at c. 5500 words alone. Perhaps my comments below will provide some scope for a reduction in length and a clearer focus. I think there is too much repetition between the results and discussion, and a stronger discussion would be one that links together the different proxies more effectively.
Line 392: Aulacoseira are also common in lakes that stratify for part of the year, e.g. during ice cover or in warmer summer months. Not all Aulacoseira species are large, but there different sizes does have implications for resource competition for e.g. dissolved silica
Line 399: “…and a seasonal ice cover does not necessarily reduce its abundance” – indeed! I don’t think diatomists would ever think this the case because the majority of lakes are dimictic.
Line 400-402: I'd agree with this statement, but changes in temperature between air and surface waters, and wind strength & direction are also very important in causing overturn, not just snow-melt.
Line 415: The authors suggest that diatom concentrations “likely indicates favourable growth conditions at Lake Khamra that enhance diatom productivity”. Possibly, but caution is needed here.
Concentrations are dependent on SARs, so if accumulation rates increase while diatom production remains constant, then diatom concentrations will decline without without needing to infer low-productivity lake. The authors know this, as elsewhere it is stated that DVC vary with SAR. Also, it looks like you are comparing DVC with an average of your whole core, with surface sediments from other cores, which will likely have higher SAR as sediments / core compaction is minimised at the surface layers.
I think it is important to make the decision early on in the study if you are going to present and discuss DVCs or DVARs. I’d recommend the latter, and DVCs do not therefore need to be discussed at all beyond the calculation being an intermediate stage to get to DVAR.
Section 4.2:
I found this section quite difficult to keep track off. There is repetition with the results section, when changes in diatom abundances are described by zone.
There are a lot of inferences being made with regard to changes in diatom relative abundances, and it's hard to decipher which of these are likely to be real or simply a function of having species expressed as relative abundances (as one species increases, others must decline etc).
Rather than having a discussion based on the zones (which is rather descriptive), a higher-level discussion that just focuses on the main trends would be better.
When I look at the stratigraphy in Fig 3 in relation to your discussion I’d recommend that authors consider:
- what are the ecology and habitat preferences of each of the taxa shown - draw on information from papers that look at their ecologies (lab, modelling, field).
- provide some summary plots on the stratigraphy: total planktonic, total tychoplanktonic; total benthic; PCA plots of axis 1 and axis 2 sample scores. These will all help with the interpretation and could help ensure over-interpretation of the percentage changes is not being done.
- As you have calculated total diatom fluxes, how might the stratigraphy look if all the major taxa were shown as fluxes. This means that changes in their abundance are independent from each other.
Lines 431-434: Not sure if this is a good comparison to make. Stephanodiscus species normally need less silica and do better in waters that are sufficient in nutrients - e.g. see classic Kilham et al. 1986 paper (and others) https://aslopubs.onlinelibrary.wiley.com/doi/abs/10.4319/lo.1986.31.6.1169
Lines 459-460: Be really careful of interpretation. I would not rely on a PCA biplot to inform competition ecology between your species. Go to source material for those species ecologies.
Lines 499-503: There is far too much conjecture here for so many reasons. The authors have not provided the evidence to support the claims being made as they are not comparing like (ecosystem) with like (ecosystem).
Lines 527-529: A lot of caution is required here when making comparison to Biskaborn et al. 2021b. for a couple of reasons:
First, the resolution of the stratigraphies in Biskaborn is *much* lower than for Khamra, and so there are major uncertainties in tying together what look to be similar patterns.
Second, their interpretation requires greater critical consideration when interpreting the data presented. The two shifts between major taxa - one around 1850 CE or so, and one in the 1970s - must be responses to different drivers.
- The shift at c. 1850 CE I think will be climate related, with the northern hemisphere broadly moving out of the cooler Little Ice Age. While many countries were industrialising at that time, you would need to provide economic evidence that this was also occurring in your region to cause acid rain. I’m not an expert in this region, but I don't think this exists, and no evidence is presented to show that it does exist. We see similar shifts in diatoms across Lake Baikal for example, that had little industry around its catchment.
- The shift in the 1970s is likely related to global warming and possibly increased reactive nitrogen from pollution. You have perhaps d15N evidence of this, although the decline is rather small.
Section 4.3: I wonder if it would be better to layout the changes in organic geochemistry in the lake / catchment system before the discussion of the diatoms. Then relate the diatoms to changes in the broader organic geochemistry environment?
Lines 571-572: The authors make the assumption that because climate is inferred to be drier at this time period, then erosional input from the catchment is reduced, but no evidence is produced for a decline in erosional material. While it may be the case that slower accumulation rates are linked to less catchment input, given the planktonic diatoms dominate the profile it could be as likely that sediment accumulation is linked to primary production in the lake, and if conditions are cooler and drier, then there may be less primary production.
Line 592: I think that Fig A5 should actually be shown in the main body of the paper.
Lines 604-606: again, be careful of over-interpreting data. At this site, ice cover, snow cover, depth of mixing etc are all going to be much bigger determinants of the amount of light the planktonic diatoms can use than changes in solar variability!
Line 669: Fig 2: It's odd having the figure for geochemistry being so far away from its discussion. I'd recommend re-ordering the figures, or reordering the structure of the discussion.
Line 670-671: The Suess effect can be detected on d13C values from c. 1850CE and the start of the industrial revolution....
Lines 687-690: This is quite a conclusion to make. If we look at the d30Si profile and e.g. A. subarctica, they both show a concurrent response at 1830 CE for example. d30Si values do not need to change very much to infer quite substantial changes in resource availability and nutrient cycling.
Lines 700-701: A statement is needed here as to what this level of (Hg) pollution is, contextual wise. Is it similar to background increases across the northern hemisphere? Or does it indicate local origin? Are the concentrations high enough to indicate major sources of pollution that could have caused the suggested impact on your lake?
Lines 703 - : I do have a problem with the use of the Mallomonas scales as part of an acidification index in this lake.
- While lakes in base-poor catchments may acidify due to acidifying pollutants, the catchment has dolomite in it, which surely would buffer against potential acidification
- if increasing Mallomonas scales are also indicative of increasing nutrients, then would freshwater acidification also occur?
- the increase in stelligera is not a diatom I recognise from an acid flora.
- pH 6.07 is not a low pH; lake acidification becomes problematic when pH values fall below a critical level of 5.6.
- Remember a correlation between Hg and Mallomonas scales does not mean cause and effect. Is the amount of Hg increasing in the lake likely to be an indicator of enough acid rain to cause lake acidification?
- A lot of this discussion of acidification is based on the conclusions drawn by Biskaborn et al. 2021b. Biskaborn show that Mallomonas scales are actually only found at depth in two of their four cores. And in the other two cores, scales are found only in the surface sediments and therefore cannot represent any acidification trend (for example in their dated PG2203). So during their 'industrial period' (their term), there are no Mallomonas scales apart from the very surface layer. In PG2208, there are scales down-core, but these start to increase before the period of 'industrialisation', and therefore cannot be attributed to acidification either.
Personally, I’m not convinced that the Briskaborn study shows (indirect) evidence of acidification at all. But regardless of my opinion on a separate study, the authors of this current study need to be more critical in the evidence presented for lake acidification.
Line 721-724: The increase in A. formosa is very small (c. 4 to 6%?) - this is likely just natural fluctuation. Moreover, it is certainly smaller than the increase around 1810 - 1820 CE. What is the explanation for the larger fluctuation here - it can't be acid rain
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 140 | 30 | 11 | 181 | 5 | 7 |
- HTML: 140
- PDF: 30
- XML: 11
- Total: 181
- BibTeX: 5
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1