the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Assessment of the seasonal cycle in Atlantic density flux
Abstract. An analysis of the seasonal/sub-seasonal cycles of density flux is now possible thanks to advances in satellite oceanography. The kinematic density flux framework, developed to infer the buoyancy-driven ocean circulation using high-resolution satellite datasets, was applied at 1/4° resolution to monthly maps of satellite-derived Sea Surface Salinity, Temperature, and Currents (SSS, SST and SSC) over 2011–2020. Combining them with a blended satellite/in-situ Mixed Layer Depth (MLD) dataset, we derived density flux estimates throughout the Atlantic. We also performed a harmonic analysis to the density flux estimates, to diagnose the contribution of thermal and haline processes to density flux. We find that the sub-tropics and mid-latitude annual cycle explains 70–80 % of the variability in net density flux. With the addition of a semi-annual cycle, the explained variance reaches 80–85 %, suggesting density flux is sensitive to other atmospheric/oceanic processes with higher/lower temporal frequencies. Haline processes dominate density flux variability in the Denmark Strait, and parts of the Labrador and Norwegian Seas – all crucial areas for the Atlantic Meridional Overturning Circulation. The subpolar North Atlantic density flux is primarily governed by haline variability, with freshwater forcing driving most monthly extremes and exhibiting a quasi-symmetric pattern of alternating positive and negative events. Anomalous thermal contributions and localized salinification in December 2020 mark a striking departure from prior years, raising the question of whether this signals a regime shift or a singular event.
- Preprint
(35134 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 26 Aug 2025)
-
RC1: 'Comment on egusphere-2025-2973', Anonymous Referee #1, 16 Jul 2025
reply
This study looks at the seasonal variability of density fluxes in the Atlantic basin between 2011 and 2020. It computes the mean, seasonal cycle amplitude and phase. It also looks at density flux in the subpolar North Atlantic.
Looking at the seasonal variability of surface fluxes of density is a valuable exercise, but I did not find this paper did that in a coherent way. This is evidenced by the proliferation of detailed comments below. I am ambivalent about whether to allow the authors to resubmit, or urging the editor to reject it. However, due to the large number of issues shown below, I would recommend rejection.
The paper is poorly focused. Figs. 1-4 are about the mean and seasonal cycle of density flux over the entire NA basin. However, much of the text focuses on the subpolar NA, and on interannual variability.
Detailed comments
7. This number "70-80%" does not appear anywhere in the paper besides here. Ditto the next line (80-85%). The comparison of annual and semi-annual cycles mentioned here is brushed off as irrelevant in the text of the paper (line 206).
16. "Surface density flux is a fundamental driver of oceanic buoyancy exchange" Density flux and buoyancy exchange are synonymous, so this is a tautology.
68. The BEC website referenced here says that the dataset is available "2011-2019". The ftp site for downloading the data was not operational when I tried to access it.
85-86. This seems a stretch, to use SSS/SST data to infer profiles down to 1500 m. I am not sure why the authors need to do this, or what the validity of such profiles would be. It depends on what these synthetic profiles are being used for - which is not evident on reading through the entire paper.
123. Why is the word "subpolar" here? This paper is supposed to be about the entire Atlantic basin.
Figure 1. The caption should state that a positive number means density gain by the ocean. Also, it appears that the ocean experiences a net gain of density- there are much more red colors on this map than blue. Shouldn't there be a balance between loss and gain? Or maybe the gain in the Atlantic basin would likely be balanced by loss elsewhere. Or if the predominance of density gain is expected, then the authors need to explain why.
162. The haline contribution at high latitude may be difficult to compute due to the uncertainty of satellite SSS in cold water. The noisy haline component at 50-60S verifies this.
168-169. It's not clear from Fig. 1 that the Amazon and Congo Rivers contribute much to the overall NA buoyancy budget. The red and blue areas are interspersed, and strongly confined to the coast.
181. The maximum heat and freshwater flux typically occurs in late boreal winter, Feb.-Mar. See Wang & Carton (2002, Fig. 5). This is especially true in the subpolar NA.
https://doi.org/10.1175/1520-0485(2002)032%3C3474:SHBOTN%3E2.0.CO;2
196-198. This is a vague generalized statement. Perhaps the authors can elaborate, or provide the reader with a reference that goes into more detail. Sea ice melt/freezing would certainly have a strong annual component, along with continental runoff. I don't know about prepcipitation, but I would guess there is a seasonal component of variability to that too.
202-205. I guess Fig. 3f is being referenced here. It's hard to tell from the figure exactly what the month of maximum density flux in the SPNA is, or even where the authors are referring to.
207. "comparatively small" Can the authors give some numbers to back this up.
216. Remove "(80%)".
216-218. This is unclear. It needs re-wording.
219. Figure 5a is referenced here, but I see no reference to Fig. 4. Maybe this refers to 4a?
Figs. 4a and b. It looks like the variance explained for the total density flux is less than that explained by the thermal part in much of the SPNA (and the rest of the Atlantic basin too). I thought that the total variance explained would be the sum of the thermal and haline parts, so the total could not be less than either component. That is what the figure caption implies, and would be the most sensible way of displaying this information.
224-225. I do not see how this indicates what the authors say it does.
225-229. None of this is evident from looking at Fig. 4. Perhaps it would be valuable to display a map of the ratio of explained haline variance to the explained thermal variance. It seems that in the SPNA, density flux has a weak seasonal cycle, but that does not imply that haline processes are important.
231. This plot, while interesting, does not belong in a paper about seasonal variability (see the paper's title).
Fig. 5. The area covered by this plot needs to be specified somewhere, with exact coordinates and a box (or region) on a map. Also, in the caption: I think the authors mean "lightening", not "lightning". Only 3 of the 4 events shown are lightening events. The first dashed line does not correcpond with one.
239-244. What are these events? I guess a positive (negative) event is indication of the SPNA salinifying/cooling (freshening/warming). They seem to occur mainly in the late fall/early winter. There must be some clues. Perhaps they are associated with advective events, or ice melt/freezing. Line 249 refers vaguely to freshwater forcing, but does not give the reader a sense that this has been carefully thought through or investigated.
254. I note that this Summary section hardly mentions the seasonal cycle of surface density flux, which is ostensibly the subject of this paper (see title). I would also note that this section barely mentions any previous results. One of the purposes of a Discussion section is to do this. There may or may not be any previous estimates of surface density flux, but there are plenty of studies of the surface thermal and haline budgets in the Atlantic at the seasonal time scale. How do the results presented here compare to those? (Almost none of these studies are mentioned in the Introduction either.)
255. I do not think this study does this at all. It seems to show that the seasonal haline component of density flux is mainly small compared to the thermal part, i.e. Figs. 3b,c, except in very limited areas. (What do the authors mean by "modulating"?)
255-257. "observed sensitivity of density flux to freshwater perturbations suggests a potential weakening mechanism" What is this referring to? Where is this demonstrated?
264-268. With respect, I generally find this sort of self-aggrandizement unnecessary and distasteful - and not warranted. This point can be made without using words like "unprecedented" and "significant advancement".
273. This repeats line 255.
276. This is a reference to a dataset, which is not appropriate in this context. A reference needs to be given to a paper that describes the approach to justifying and creating the dataset.
277. I do not see how it emphasizes this. Numerical models are not used or even mentioned in this paper. Even the in situ product used for estimating MLD is not model-based (lines 83-89). (I searched for the word "model" in the manuscript and found it only in the "Summary and Conclusions" section.)
287. "unprecedented accuracy" If this is truly the first product of its kind (line 280), these words are unnecessary.
318. The Worsfold & Martin reference is not in the reference list.
354-355. A URL or DOI is needed where the data can be accessed.
Citation: https://doi.org/10.5194/egusphere-2025-2973-RC1
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
48 | 15 | 3 | 66 | 2 | 2 |
- HTML: 48
- PDF: 15
- XML: 3
- Total: 66
- BibTeX: 2
- EndNote: 2
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1