Wind-driven transport of fresh shelf water into the upper 30&thinsp;m of the Labrador Sea

Schulze Chretien, Lena M.; Frajka-Williams, Eleanor

doi:https://doi.org/10.5194/os-14-1247-2018

Articles | Volume 14, issue 5

https://doi.org/10.5194/os-14-1247-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/os-14-1247-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 14, issue 5

Research article

|

15 Oct 2018

Research article |

| 15 Oct 2018

Wind-driven transport of fresh shelf water into the upper 30 m of the Labrador Sea

Lena M. Schulze Chretien and Eleanor Frajka-Williams

Download

Final revised paper (published on 15 Oct 2018)
Preprint (discussion started on 07 Mar 2018)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of Wind-driven transport of fresh shelf water into the Labrador Sea Basin', Anonymous Referee #1, 04 Apr 2018
- AC1: 'response to reviewer 1', Lena Schulze, 29 May 2018
RC2: 'Review of Wind-driven transport of fresh shelf water into the Labrador Sea Basin', Anonymous Referee #2, 09 Apr 2018
- AC2: 'response to reviewer 2', Lena Schulze, 29 May 2018

Peer-review completion

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

AR by Lena Schulze Chretien on behalf of the Authors (29 May 2018) Manuscript

ED: Referee Nomination & Report Request started (23 Jun 2018) by Matthew Hecht

RR by Anonymous Referee #2 (31 Jul 2018)

Suggestions for revision or reasons for rejection

Overview
- - - - - - - - - - -

In this revision the authors have made many improvements to the manuscript, addressed most of my concerns from the previous review and substantially improving the quality of the writing. I have included an additional list of comments and questions below.

My major outstanding concern is that this article, as currently presented, is going to lead other scientists to conclude that the transport of freshwater into the LS is principally wind driven. In fact what the authors actually demonstrate is that the freshwater transport into the top 30m of the LS is principally wind-driven, i.e. that the transport within the Ekman layer is principally wind-driven. The authors calculate that 60% of the freshwater transport in the top 100m is wind-driven, and that eddies “become more important” when the calculation is extended to 200m. Therefore, by the authors’ own calculation it seems to be that the most reasonable conclusion is that eddies are still the most important mechanism of freshwater transport into the LS.

Based on the authors’ results, we can reasonably draw the following conclusions:
1. A component of the freshwater inflow into the LS is controlled by Ekman transport.
2. This component occurs is largely confined to the upper 30m of the water column.
3. This component is smaller (though the authors have not quantified by how much) than the total freshwater transport into the LS by eddies (over all depths).
This is a crucial distinction that is not adequately conveyed by the title, abstract and body of the manuscript. The authors’ failure to make this distinction has already caused confusion - in their first review, reviewer 1 made the the following remark:

“Since the authors are going to dispute the commonly held paradigm that eddies are the
main exchange mechanism from the WGC, …”

Again, by the authors own admission, “eddy fluxes become more important only when extending the calculation to 200m”, so clearly they cannot be disputing the paradigm that eddies are the main mechanism of transport into the LS. Yet this is what reviewer 1 took away from this manuscript, and presumably what many other readers would too.

My major recommendation is that the authors change the title, abstract and body of the manuscript to remove the ambiguity as to their results, i.e. that winds control a relatively small component of the freshwater flux confined to the upper 30m. For example:
1. A suitable title would be “Freshwater fluxes into the upper 30m of the Labrador Sea are dominated by wind transport”.
2. The authors’ statement in the abstract that “60% of the top 100m enters the basin in the top 30m along the eastern side” is accurate but misleading, because they reveal in the manuscript that eddies dominate the freshwater flux over the top 200m.
3. The last sentence of the abstract is also misleading: “the year-to-year variability in the freshwater transport … is dominated by wind-driven Ekman transport, rather than eddies”. This is based on the Lagrangian trajectory analysis, and only applies to the top 30m of the water column - there is no reason to think this would still be true if the analysis were extended to 200m.
4. At the end of the manuscript, summary conclusions are drawn without explicitly acknowledging that they apply only to the top 30m of the water column.

As I said in my previous review, I agree with the authors’ final statement of the article, which takes a more modest perspective on the results: “in a region where the freshest water is concentrated at the surface and winds are strong, the surface Ekman transport cannot be neglected”. I urge the authors to present the rest of the article with a similar perspective.

Comments/questions:
- - - - - - - - - - -

NOTE: Page and line numbers refer to the tracked changes version of the article.

Regarding variance in crossing probabilities: in their response the authors state that the “Ekman transport variations explain more than 70% of the variance in the particle crossing probability”. Table 2 gives an r-value of 0.72 for this correlation in the northeast, and % of variance explained is r^2 = 51.8%. In the southeast it’s more like 25% of variance explained.

The “peaks” issue: I still take issue with the authors’ claim that there are two distinct “peaks” of freshwater inflow in to the (top 30m of) the LS. It is not sufficient to simply judge by eye from Fig. 6 that two peaks exist - clearly I and the authors have reached different subjective conclusions this way, and an objective method of distinguishing “peaks” is needed.

The authors note that a peak can occur due to a small amount of very fresh water entering the basin, rather than a large volume of somewhat fresh water. Qualitatively, I follow this argument, but the freshwater “peak” in September is actually saltier than in March/April, AND is associated with a 5x lower crossing probability (Fig. 5b,d).

The authors note that there is a large influx of salty water in the Southeast during the March/April peak (see also L532-535), but without some kind of quantification of the relative magnitudes of these fluxes it is impossible to draw objective conclusions. For example, is the salt influx in the Southeast so large that we shouldn’t even consider the March/April freshwater inflow in the Northeast to be a “peak” any more?

The authors state that they “have estimated a freshwater flux from the number of particles that cross into the basin”, but that they “did not feel the calculation warranted publication”. Again, if the authors cannot defend their conclusions quantitatively then I argue that they should not be drawing those conclusions in the first place.

L224-225: My previous comment about topographic form stress was intended as a correction: friction is not the only process that extracts momentum from the fluid at the sea floor - bottom form stress does too. Also, how is bottom friction represented in this model? Is there a simple quadratic drag law, or does the model actually simulate vertical mixing/viscosity in the bottom boundary layer?

L251: “represents” -> “represent”

L265: “study” -> “study”

L266-267: I don’t understand what “drastically developing” means - could the authors be more specific.

L275: “high,” -> “high:”

L293: “in many studies” is not a substitute for citations. I specifically highlighted this omission in my previous review, and am surprised to see that it has not been rectified.

L298: There is a double parenthesis at the start of the citation on this line.

L306: Citation to Luo et al. should not be in parentheses.

L325: Citations in a sentence should form a comma-separated list, rather than a semicolon-separated list.

L329: “inshore the” -> “inshore of the”

L468: “north-“ -> “northeast”

L481–482: Is the crossing speed consistent with the Ekman velocity?

L551: Are these transports correct? A few mSv seems very low for Ekman transport across such a long section of the basin edge. I would have expected something on the order of a few Sv.

L571: “water” -> “waters”

L571: “probability” -> “probabilities”

L633: “amount” -> “probability”

L635-642: It is interesting that the shelves become saltier during times of low freshwater flux, yet the EKE remains approximately constant. A mixing length scaling for an eddy diffusivity of freshwater would suggest a diffusivity that scales with EKE^(1/2), and the freshwater flux is diffusivity * freshwater gradient. So during times when the shelves are fresher, we would expect stronger eddy freshwater fluxes into the LS because there’s a larger freshwater gradient between the LS and the shelves, but more-or-less constant EKE and thus diffusivity.

L695: I am struggling to discern how the authors have concluded that the second pulse is “about three times stronger than the first pulse”.

L729—731: Is r=0.72 “remarkably high”? I guess with just over 50% of variance explained it is (barely) justified to say that Ekman transport plays the “primary role in the variability of freshwater transport”, but the authors may be over-selling it here.

L735-736: … in the upper 30m.

L755-763: In their response the authors acknowledge that this calculation of the Ekman freshwater flux will be sensitive to the choice of reference salinity. One could contrive choices of this salinity that make the Ekman freshwater transport equal to zero or that make it larger than the total Ekman volume transport. Please provide some measure of this sensitivity, i.e. take a range of reasonable reference salinities and provide the Ekman freshwater flux as a corresponding range.

L762—763: Please quantify the relative importance of eddy and Ekman transports over the top 200m.

L766—768: … in the upper 30m.

L768—769: … in the upper 30m.

L772–773: … in the upper 30m.

Hide

ED: Publish subject to minor revisions (review by editor) (03 Aug 2018) by Matthew Hecht

AR by Lena Schulze Chretien on behalf of the Authors (13 Aug 2018)

ED: Publish subject to technical corrections (07 Sep 2018) by Matthew Hecht

AR by Lena Schulze Chretien on behalf of the Authors (07 Sep 2018) Author's response Manuscript