Components of 21 years (1995–2015) of absolute sea level trends in the Arctic

Ludwigsen, Carsten Bjerre; Andersen, Ole Baltazar; Rose, Stine Kildegaard

doi:https://doi.org/10.5194/os-18-109-2022

Articles | Volume 18, issue 1

https://doi.org/10.5194/os-18-109-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

Towards an integrated Arctic observation system to fill gaps...

https://doi.org/10.5194/os-18-109-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 1

Research article

|

21 Jan 2022

Research article |

| 21 Jan 2022

Components of 21 years (1995–2015) of absolute sea level trends in the Arctic

Carsten Bjerre Ludwigsen, Ole Baltazar Andersen, and Stine Kildegaard Rose

Download

Final revised paper (published on 21 Jan 2022)
Preprint (discussion started on 05 Nov 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of “Assessment of 21 years of Arctic Ocean Absolute Sea Level Trends (1995-2015)” by Ludwigsen et al.', Anonymous Referee #1, 15 Dec 2020
- AC1: 'Reply on RC1', Carsten Ankjær Ludwigsen, 26 Feb 2021
RC2: 'Review of "Assessment of 21 years of Arctic Ocean Absolute Sea Level Trends ..."', Anonymous Referee #2, 17 Dec 2020
- AC2: 'Reply on RC2', Carsten Ankjær Ludwigsen, 26 Feb 2021

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Carsten Bjerre Ludwigsen on behalf of the Authors (26 Feb 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (02 Mar 2021) by Anne Marie Treguier

RR by Anonymous Referee #3 (13 Mar 2021)

Suggestions for revision or reasons for rejection

This revised version has neen significantly improved compared to the original manuscript. However it still needs substantial improvement before being considered as publishable. My main comments concern : (1) the writing, sometimes poor and unclear, with several vague or not argued statements (see below), (2) the lack of discussion on the data uncertainties, in particular on the steric component, and (3) the lack of acknowledgement to similar previously published studies on arctic sea level (just quoted in the reference list but without discussion/comparison) (see below).
- The abstract is vague and non informative. Nowhere in the abstract is explained what the computed reconstructed sea level consists of. The abstract says : ’as the first study…’, a wrong statement in view of the many previously published studies. In addition, it claims that NOT using GRACE is a progress, but why ? GRACE has a nearly global coverage over the Arctic and provides unique information about tha mass component of the sea level budget in the Actic. Why oppose GRACE to the approach considred here ? The only acceptable argument is the shorter time series of GRACE data. I recommend to rewrite the abstract and clearly explain the approach considered in this study and presensing the main outcomes (and eventually the novel results compared to previous studies). More detailed comments below :

- In the introduction, describe the main objective of this study and the approach used
- Line 31 : what is the meaning of ‘difficult’ in the sentence ‘Previous attemps in reconstructing sea level in the Arctic have show to be difficult’ ? Explain. In the same sentence, the author write : … because satellite and in situ observations are less consistent’. Less consisent than what ? Curious claim considering that the authors ALSO use satellite and in situ observations…
- Line 35 : Clarify what you want to say by : ‘the 10 mm/yr discrepancies between GRACE solutions’. In absolute value, this has no meaning. It should be compared to the amplitude of the signal. It also depends where.
- Line 37 : I disagree with the sentence ‘ some authors tend to choose the GRACE solution that closes the sea level budget. This not true. Please remove this sentence.
- -Method section, lines 63-64 and Equation 7 : What about local VLMs , unrelated to GIA and ongoing mass redistribution ? These could be important at some TGs
- Eqution 6 : What about the GIA component for altimetry (the component of -0.3 mm/yr removed to altimetry data in terms of global mean)? It seems to be ignored here.
- Line 90. Fig. 1 should be quoted first.
- Section 3.1, Altimetry. What is the accuracy of the altimetry data set in this region partly covered by ice. The authors disgard GRACE because of its ‘poor’ accuracy, but what about altimetry ?
- Section 3.1, VLMs. The authors choose the option of using theroretical VLMs due to GIA and ongoing mass redistributions rather than GNSS data. Ok but explain why using the model is best and at least compare where possible the model VLMs with GNSS.
- 3.3, steric sea level. The interesting part of this study is the use of situ T/S profiles over the Arctic. However, we have no idea of the data coverage nor on the accuracy of this data set. What is the integration depth H ? I strongly recommend the authors provide information on the T/S data set, a crucial aspect of the study.
- Section 3.4, ocean mass component. This section is hard to follow, with several unclear sentences, e.g., lines 163-165.
- Results section. I strongly recommend to put into perspective the results of this study with the published literature (e.g., the works by Henry et al, Armitage et al., Carret et al., Raj et al). What is novel ? Wht is different ? etc.

Referee Report: PDF

Hide

RR by Anonymous Referee #2 (19 Mar 2021)

Suggestions for revision or reasons for rejection

Second review of Ludwigsen et al.

When I reviewed the first submission of this manuscript, I recommended rejection. I cannot even remember if I have ever before made this recommendation and I have reviewed a great many manuscripts over the years. My assessment then was this: "this work reads like a technical report that has not properly matured into a scientific study, and I do not think that it merits publication". I now repeat this recommendation.

I started reading this second, revised, version, and I observe that the Introduction is very short (LL 16-43) and makes some (referenced) assertions about "uncertainties" (L 24) and "discrepancies" (LL 34-5), mentions that Arctic sea level reconstruction is "difficult" (L 31) and says that "satellite observations and in-situ observations are less consistent than in low and mid-latitudes"; but none of this provides a clear and explicit rationale and context for the work that is described in the rest of the manuscript.

The requirement on the author of any manuscript is to state in the Introduction (or Background, or Rationale) the following: (1) what is the "big issue", (2) what do we know, (3) what do we not know, (4) the reason for the importance of our ignorance, and (5) what we propose to do about it. If I remember right, the best-known (most accurately measured) quantity in physics is the electron magnetic moment, relative standard accuracy of 3x10^-10: everything is uncertain, and quantities in the environmental sciences may be very uncertain; but does that uncertainty matter? Is it important? Why? There is enough published in the Arctic context on altimetry, gravimetry, tide gauges, land movement (and so on) for the authors to fill in thoroughly on what text is needed to embrace all of these five points, and to make the case for why their work was worth doing. I strongly suspect that it was worth doing: but the authors still need to state their reasoning clearly, explicitly, and with evidence. A similar criticism applies to the Conclusions: what new have we learned, above and beyond what we already knew from previous studies? At present, the Conclusions are a summary but do not state what is new. This is why I stated in my first review that the manuscript resembled a technical report, and that assessment has not changed.

This manuscript is still not a mature document, and I consider it wrong first to send a document in such a condition to a journal, and then to expect the reviewers and the editor to repair its deficiencies. It is the authors' responsibility to think for themselves, to seek advice from colleagues, and to consult online resources, if that is what's required, in order to present a mature document for consideration for publication. Then the reviewers will engage willingly, so that it can be checked in the usual terms, and the process not used as a substitute for manuscript development.

Hide

RR by Anonymous Referee #1 (27 Mar 2021)

Suggestions for revision or reasons for rejection

Second review of “Assessment of 21 years of Arctic Ocean Absolute Sea Level Trends (1995-2015)”

The authors have made substantial changes to the structure of the manuscript and some figures and it is now significantly clearer to read and understand. In particular, the motivation in the introduction is more coherent and I appreciate the care taken to clarify the naming of different terms. In general, my previous comments have been addressed and I think this is an interesting study worthy of publication. I have a few remaining questions/comments but these are generally very minor and I have outlined them by line number below.

Note that there are very minor typos in the manuscript (generally related to plurals) that I have not listed but should be picked up during the final proofing.

Line-by-line comments:
Line 5: “that is independent from any observed sea level change”. This is a little confusing, since the reconstruction is based on observations. What is independent from what? By the observed sea level change here, do you mean tide gauges? Some clarification would be welcome

Line 73: Why is GRACE listed as a satellite product that is used in the study?

Line 78: should there be a dot above ASL_A? In general, I recommend checking the variables from equations that appear in the main text for this. Also check equations 8 and 9

Line 130: I may be wrong, but I think the Armitage and Davidson (2014) reference here, and all subsequent references to that paper, should actually be referencing the Armitage et al (2016) paper in the reference list

Line 163-165: these sentences are confusing. Please reword, and refer to the subpanel of figure 4 that the terms discussed are related to

Line 173: this is the first time ECCO has been introduced. A bit more information about this product would be welcome, especially as further on the discrepancy between ECCO and the manually-calculated Mdot are discussed

Line 177: Figure 5 should be referred to here in relation to ASL_r

Line 179: Refer to figure 4f here

Line 192: has anyone else found such a difference before? Are there any references for the sort of magnitude difference you could expect from comparing two such datasets?

Line 207: this dataset is from 2003-2014. Here it should be explained to the reader why the altimetry-based dataset of Armitage et al (2016) does not suffer from the seasonal bias and ‘flattening’ you describe.

Lines 199-207: This paragraph focuses on key differences in the Beaufort Gyre region, but there is no mention of the paper by Giles et al (2012) who specifically focus on trends from satellites in the region. This paper should be discussed here.

Line 217: specify the interior Arctic Ocean, or it suggests TGs have no use in the study

Lines 225-226: which error estimate are you talking about here? You state the tide gauges, but it is ASL_A that has the lowest error estimates for the specified locations, not ASL_TG…

Line 233: please reference figure 3 here

Line 242: I am surprised that this section does not refer to Figure 2. For these four stations in particular, the timeseries in Figure 2 is variable and one could argue that at Izvestia Tsik leaving out the last year or two of the timeseries would significantly alter the linear trend, while Golomianyi has a lot of missing data towards the end which may skew it upwards. Some acknowledgement of this figure would be welcome here

Line 242-244: it may be worth reminding the reader why the manometric estimate from ECCO and the summed estimate could differ here

Line 294: there have been studies of

Line 297: what was the time period used for the previous correlations? Could that have affected the results? This should be stated here as it is important for the conclusion

Conclusion: Given it is the main motivation for the study, I feel that a sentence or two in the conclusion could be dedicated to emphasising the novelty of this study and how it does/does not improve on the GRACE-based results in the literature. At the moment, a sentence on an improved correlation is given but is then followed by a discussion of uncertainties, which dampens the message. Perhaps moving the first two sentences of the second paragraph of the conclusion to the end of the first would help

Figure 4: the caption says “pherical glaciers” which I think should be “peripheral glaciers”? Also, I am a bit confused by how c) and d) look so different but have a very similar yearly contribution

Figure 5: is there a reason that the colormap has a different scale to figure 3? Providing both figures on the same scale would make the relative contribution of halosteric sea level change easier to see

Technical corrections
Line 20: “sea ice floats” -> “sea ice floes”?

Line 112-113: “and therefore does not significantly affect the trend estimates with a seasonal bias” -> “and therefore the trend estimates are not significantly affected by a seasonal bias”

Line 125: “does all TGs show” -> “all TGs show”

Line 129: “300.000” -> “300,000”

Line 130: “spatial and temporal” -> “spatially and temporally”

Line 134: “which is cause to large uncertainties” -> “which results in large uncertainties”

Line 150: brackets are unclear. Do you mean “…changes in ice loading (equation 4, similar to the elastic VLM-component) is computed…”?

Line 181: “way smaller” -> “much smaller”

Line 184: “til” -> “still”; line 185: “a absolute” -> “an absolute”

Line 202: “i.e.” -> “e.g.”

Line 203: erroneous bracket to be removed

Line 239: “Left map” -> “The left map”

Line 319: “thus can the individual contributions” -> “thus the individual contributions can”

Hide

ED: Reconsider after major revisions (29 Mar 2021) by Anne Marie Treguier

AR by Carsten Bjerre Ludwigsen on behalf of the Authors (08 May 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (10 May 2021) by Anne Marie Treguier

RR by Anonymous Referee #1 (02 Jun 2021)

RR by Anonymous Referee #2 (10 Jun 2021)

Suggestions for revision or reasons for rejection

Third review of Ludwigsen et al.

In my second review, I stated my dissatisfaction with the presentation of an immature manuscript and recommended rejection. I now review this third version, and I still do not find it to be publishable. Nevertheless, it is now at least in a reviewable form, so I will recommend "major revision" and state my reasons below.

The Abstract remains a list of things the authors have done; there is no statement of aim(s). Extension of record length is useful but not very interesting by itself. The authors have a lot to say about the presumed deficiencies of GRACE products (fair enough) - so what, then, is their GRACE-related aim?

They generate new, separate records for changes in SSH due to steric changes and to mass distribution changes. The mass distribution changes rely fundamentally on tide gauge records. Tide gauge records are (by their nature) coastal. A good fit to coastal sea level is not evidence of good fit over the whole Arctic domain, however; at best it might be considered relevant to the shallow shelf seas. The reason for interest in GRACE in the Arctic is that it provides data on the deep, central part of the ocean. The authors cannot (as they do at the end of the manuscript) claim that their fields can validate GRACE when their fields are not validated away from the coast. There does exist a third resource apart from GRACE and their fields (independent, therefore) in the form of long-term bottom-pressure recorders in the Beaufort Gyre. They should use these (there may be more, maybe one in the North Pole observatory, I'm not sure about that) to check on their product in full ocean depth away from the coast.

There are two other good papers by Peralta-Ferriz (GRL 2010 and 2011) that have a lot of insightful material in this regard. Imagine a mode of SL variability that is like a drum resonance - pinned around the edge and moving up and down in the middle. Tide gauges do not see it. P-F and GRACE do. This is relevant to the Ekman spin-up / spin-down mechanism (Proshutinsky Phil Trans 2015, originally Proshutinsky & Johnson JGR 1997).

So what they need to do to make this manuscript acceptable is follow up on their introductory assertions about problems with GRACE near land and actually quantify the improvements they claim for their method by first comparing their fields with BPRs as an independent check, and then comparing the GRACE mass fields with theirs. GRACE products are readily available, so show the differences between their fields and GRACE. Is it true that they can demonstrate improvement near the coast? By how much? Is it significant? What about over the deep ocean where they tide gauge validation is not relevant? If their product is relatively unconstrained over the deep ocean, is GRACE actually better there?

Answer these questions satisfactorily and they may have an interesting, useful and properly validated study.

One detail: "However, the significant change in the Beaufort Sea coincides with the transition from Envisat to CryoSat-2 and a inter-satellite bias in DTU/TUM Altimetry can not be excluded". Armitage (JGR 2016) treats Envisat / Cryosat crossover in the supplementary material, so their statement is false.

Hide

ED: Reconsider after major revisions (15 Jun 2021) by Anne Marie Treguier

AR by Carsten Bjerre Ludwigsen on behalf of the Authors (07 Oct 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (15 Oct 2021) by Anne Marie Treguier

RR by Anonymous Referee #2 (21 Oct 2021)

ED: Publish subject to minor revisions (review by editor) (30 Oct 2021) by Anne Marie Treguier

AR by Carsten Bjerre Ludwigsen on behalf of the Authors (26 Nov 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (01 Dec 2021) by Anne Marie Treguier

AR by Carsten Bjerre Ludwigsen on behalf of the Authors (08 Dec 2021) Manuscript

Short summary

This study uses a novel satellite-independent approach to quantify the components of Arctic sea level change. The 21-year time series allows studying climate-related changes in Arctic sea level. The decomposition shows that fresh water is governing sea level change, while Arctic land ice loss contributes to a small Arctic sea level rise. The reconstruction yields good agreement with sea level observations from altimetry, despite both datasets being challenged by the harsh environment.