Multidecadal sea level rise hiatus in the tropical Atlantic margin off northwest Africa

Ibrahim, Hamed; Sun, Yunfang

doi:10.5194/egusphere-2025-2546

Preprints

https://doi.org/10.5194/egusphere-2025-2546

Preprints

26 Jun 2025

| 26 Jun 2025

Multidecadal sea level rise hiatus in the tropical Atlantic margin off northwest Africa

Hamed Ibrahim and Yunfang Sun

Abstract. Satellite and reanalysis data sets are used to find and explain the drivers of a multidecadal sea level rise hiatus in the tropical North Atlantic margin off northwest Africa. The study domain sea level was rising as far back as 1986, and the hiatus began around 2010 and stopped in 2019. Mean sea level anomaly during a rising period (1996–2004) was compared to the hiatus period (2010–2018). Results show that domain-wide seawater expansion owing to shifts in density structure (steric shifts) contributed 74 % of the multidecadal sea level shift and increase in mass contributed 22 %. There are, however, regional differences in the steric and mass shift patterns. In the northern subdomain, the steric shift is dominated by salinity-driven (halosteric) expansion, whereas in the southern subdomain the steric shift is dominated by temperature-driven (thermosteric) expansion. Stronger anticyclonic circulation shift in the Guinea Dome, a permanent upwelling region where isotherms are displaced upwards, and the necessary adjustment of horizontal flow toward this anticyclonic sink, enable accumulation of low-salinity water in the northern subdomain and precipitation in the southern subdomain. The low-salinity water influx to the northern subdomain is linked to a shift in the southward-flowing Canary Current. This current was freshened by subpolar North Atlantic waters that reached the northwest African coast via two pathways: an open ocean path that is consistent with the Azores current, and a Western Europe coastal ocean path. These results highlight a multidecadal linkage between subpolar salinity anomalies and tropical sea level anomalies in the North Atlantic, with a transit period of about 5.5 years.

Received: 30 May 2025 – Discussion started: 26 Jun 2025

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 paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 4311 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (4311 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

01 Jul 2026

A multidecadal sea level rise and its hiatus in the tropical Atlantic margin off northwest Africa

Hamed D. Ibrahim and Yunfang Sun

Ocean Sci., 22, 2059–2081, https://doi.org/10.5194/os-22-2059-2026,https://doi.org/10.5194/os-22-2059-2026, 2026

Short summary

Hamed Ibrahim and Yunfang Sun

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-2546', Anonymous Referee #1, 06 Aug 2025

This manuscript aims at understanding the causes of the multidecadal sea level rise hiatus that occurred in 2010-2019 in the eastern north tropical Atlantic margin off northwest Africa. To do so, it mainly uses the ocean reanalysis ORAS5, performs an empirical orthogonal function analysis to define subdomains, and decomposes the sea level anomaly into its different components.
This manuscript is interesting, of good scientific quality, well structured, and well-written making it easy to follow the stream of thoughts. It addresses the scientific question of better understanding the drivers of sea level variability in the margin off northwest Africa and how it could be linked to remote processes.
I have two major comments and a few minor remarks which I think should be addressed before publication.
Major comment:
My first major comment concerns the scientific question being asked and which is not aligned with the analyses and results. It seems to me that the analyses presented do not answer the question aimed by the authors (specified lines 43-45) which is to find why the sea level stopped rising over period 2. Instead, by comparing the multi-year mean sea level and its components between the two periods, the authors answer a different scientific question which is: Why is the sea level higher during period 2 than during period 1? The concluding remarks section and the results summarized in the abstract indeed highlight the causes of the higher SLA during period 2 compared to period 1: mostly a halosteric expansion in subdomain A and a thermosteric expansion in subdomain B. The mass of the domain is also greater during period 2 than during period 1 mostly because of net precipitation in subdomain B and a net inflow of freshwater in subdomain A.
In my understanding, the analyses that would have answered the question asked by the authors (which is: why the sea level stopped rising during period 2?) would have been to study separately the two periods and their drivers and then compare them. For example, for the whole domain, the positive sea level trend during period one (fig 4a) seems to be due to a positive steric trend (fig 5a) and a positive manometric trend (fig 4b). On the other hand, during period 2, the stabilisation of the SLA (fig 4a) seems to be due to a positive manometric trend (fig 4b) being counterbalanced by a negative steric trend (fig 5a) which is itself due to a negative thermosteric trend (fig 8a). Therefore, we can interpret the pausing in sea level rise during period 2 as a result of a cooling of the domain.
I suggest that the author adjust their narrative so that the results match the scientific question being asked by either modifying the scientific question or the results presented.

My second major comment is about the potential physical linkage between subdomain A salinity and salinity in the western subpolar north atlantic. In my opinion, the manuscript does not provide enough evidence of the existence of the path 1 and path 2 and the westward flow connecting path 2 and 1. It is not clear to me, to what extent the paths 1 and 2 are already known (please provide citation if so) or if they are an original finding of the authors. These paths could be valid but should be presented with more care (as hypotheses or as potential connections) if they are not demonstrated more convincingly. It would be possible to prove a direct link for example by using water parcels lagrangian backwards tracking in time with particles released in the region of interest.

Minor comments:
Throughout the whole manuscript:
Usage of the word ‘upward’. I found a bit confusing the usage of ‘upward’ to refer to the positive shift in sea level between the two periods. The instinctive understanding of ‘upward’ usually refers to something moved higher in depth (it is used as such lines 9, 29, and 30). I recommend replacing it by ‘positive’ when referring to a shift when values are greater during period 2 than period 1.
Usage of word ‘shift’: This word is overly used and sometimes its meaning is not clear. I suggest replacing it with words like ‘change’, ‘increase’, ‘decrease’ etc. when relevant.
The word ‘hiatus’ is used very early in the manuscript but is never really defined. It is only line 45 than the reader guess than it means ‘pause in the sea level rise’. I recommend defining it early in the introduction. In addition, if the authors decide to modify the scientific question of the manuscript following my major comment, I recommend to not use the word ‘hiatus’ at all as it can be easily replaced by ‘pausing in sea level rise’ or other simple words.

All figures showing time series: It would greatly help the reader to highlight period 1 and 2 on all time series either by lightly color-shading the background or delimiting them using vertical lines.

line 5: The steric and mass contributions do not sum up to 100% which can be confusing as the reader usually expects it to do so. The reader later understands why (table A1) but I suggest clarifying this in the abstract.
Line 20-22: please add a citation to support this sentence.
Line 30: ‘Fig 10’. It seems that the figure number is incorrect. Please verify.
Line 35: ‘temperature shifted upward’. Does the author mean ‘northward’? or increase in temperature? Please clarify.
Line 81-84: It is not clear to me how the vertical land displacement is not taken into account in the study. Is it simply neglected? Is it included in the ORAS5 reanalysis or not? How does this impact the comparison with the altimetry data and grace data?

Typos and misspelling:
First paragraph: spaces are missing in units presentation. This would increase readability.
Line 83: tide gauge is misspelled ‘guage’.
Line 162: ‘gauge’ river is misspelled.
Figure 11: ‘and’ is duplicated in the caption.
Line 379: ‘here’ to replace by ‘there’

Methods:
Line 102: it is possible to simplify the sentence by removing ‘taking the difference between two satellite measurements (i.e.,’.
Line 111: replace ‘bathymetry’ by ‘seabed’ or ‘ ocean floor’
Line 153: The agreement is good between ORAS5 G and Grace data for most of the time series but not in 2003-2005 and from 2015. It would be nice to have some comments on this especially because these years are part of periods one and two.
Figure 3: panels a1 and a2 are not referred nor commented on in the text. What is the relevance of these panels?

Results:
Line 202-203: Can you elaborate more to explain why the interpretation is consistent with the SLA pattern in the EOF analysis? This is not evident to me.
Lines 208-220: These paragraphs were a bit difficult to read. I suggest reorganizing them to separate the interpretation for domain A and for domain B.
Figure 7: I am wondering why analysing the temperature and salinity instead of analysing directly the thermosteric and halosteric sea level. Is it because it is not possible to compute the thermosteric and halosteric sea level for the four layers?
Figure 9 and 12: It would help the reader to place the domain, subdomain A, and subdomain B on these maps when they are part of the plot to help to locate them. On figure 9 a1, it would also help to have a schematic arrow to represent the guinea done cyclonic circulation.
Line 273-274: It is said that layers 2 and 3 have ‘comparatively large salinity shifts’. I don’t understand. It is compared to what? To period 1? To the other layers?
Figure 10: panel a. It would be useful to show the time lag corresponding to the maximum correlation shown on panel a. This would strengthen the hypothesis of the potential pathways to link the subpolar North Atlantic and domain A.
Line 281-282: the currents located around 45N that would link path 2 to path 1 is in my opinion not visible from fig 11a.
Figure 10b: I am wondering about the relevance of this map of regression coefficients. Are the regression coefficients computed with a time lag (I understand it is not but please clarify)? I also noticed that it is barely used in the manuscript. To me the paths drawn on top of the map are not hinted at by the map itself. In other words, in my opinion, this map does not add any evidence to the possible existence of the paths indicated by the arrows. In addition, from a methodological point of view, it is not clear to me why the linear regression during period 2 between salinity averaged in domain A and the other grid points in the Atlantic would prove a causal link between the two, especially because (as it is said in the manuscript) it would take some time for the salinity in one place to be reflected in the other.
Line 287-288: Could you provide a justification to why the purple box chosen in the western subpolar north Atlantic is representative?

Line 291 and following: I am not particularly knowledgeable on potential vorticity. Is the conservation of potential vorticity a common way to try to trace water masses? If so please provide citations.
Line 303-305: To me the paths 1 and 2 are not visually clear in figure 11b.
Line 307-308: How is this lagged correlation computed? Is it from the time series of figure 7 a3? In that case it is natural that this is consistent with figure 7 a3.

Section 5: It seems that for the demonstration of this section the vertical velocities are very important but they are not analysed. Why not also analyse the vertical velocities?
Line 361-364: the shift of the northern edge of the countercurrent is not clear to me.Maybe it would be useful to indicate on the figure.
Line 364: ‘vertical velocity’. The shift of the vertical velocities is not clearly demonstrated but inferred. Please correct the wording.
Concluding section:
Point 5: Can you elaborate more on this point? Do you have examples of work that has been done using numerical simulations in that region or elsewhere? Can you hint at what model would be a good option to do this?

Citation: https://doi.org/10.5194/egusphere-2025-2546-RC1
- AC1:
  'Reply on RC1', Hamed Ibrahim, 19 Sep 2025
  
  Dear Reviewer,
  Thank you for reviewing our manuscript and for providing us with very insightful comments. We have responded to all the comments as documented in the attached pdf file.
  Regards,
  Hamed,
  
  Citation: https://doi.org/10.5194/egusphere-2025-2546-AC1
  - AC4: 'Reply on AC1', Hamed Ibrahim, 19 Sep 2025
    
    Revised manuscript for your reference.
    
    Citation: https://doi.org/10.5194/egusphere-2025-2546-AC4
- AC5: 'Reply on RC1', Hamed Ibrahim, 27 Sep 2025
  
  Dear Referee,
  In light of comments from the handling editor. We have submitted a new revised manuscript and a new track changes version. Please kindly refer to these new versions in our reply to you. We have attached this new track changes version here for your convenience. Thank you.
  
  Citation: https://doi.org/10.5194/egusphere-2025-2546-AC5
RC2:
'Comment on egusphere-2025-2546', Anonymous Referee #2, 08 Aug 2025
This study provides a valuable analysis of the 2010–2018 sea level rise hiatus off northwest Africa using satellite/reanalysis data (1993–2018). The identification of steric dominance (74%) and subpolar-tropical salinity linkages is novel and impactful. Methodological rigor and validation with independent datasets are key strengths. However, clarification of physical mechanisms and data limitations is needed for full acceptance. I recommend acceptance after minor revisions.
Eq. 6 oversimplifies vertical-horizontal coupling. Could you show decadal trends of vertical velocity (w) in the dome core using ORAS5 data?

Please discuss how anticyclonic circulation modulates downwelling against background upwelling in more detail.

While mentioned in Section 2.2, error quantification is missing. Could the authors add error bars in Fig. 4?

The 2010–2018 hiatus (Fig. 2a) coincides with subdomain A freshening (Fig. 7a2) but the causal sequencing is unclear. Please add lead-lag correlation analysis between subpolar salinity and SLA.
Citation: https://doi.org/10.5194/egusphere-2025-2546-RC2
- AC2:
  'Reply on RC2', Hamed Ibrahim, 19 Sep 2025
  
  Dear Reviewer,
  Thank you for reviewing our manuscript and for providing insightful comments that have improved our work. We have responded to all the comments as documented in the attached pdf file.
  Regards,
  Hamed
  
  Citation: https://doi.org/10.5194/egusphere-2025-2546-AC2
  - AC3: 'Reply on AC2', Hamed Ibrahim, 19 Sep 2025
    
    Revised manuscript for your reference.
    
    Citation: https://doi.org/10.5194/egusphere-2025-2546-AC3
- AC6: 'Reply on RC2', Hamed Ibrahim, 27 Sep 2025
  
  Dear Referee,
  In light of comments from the handling editor. We have submitted a new revised manuscript and a new track changes version. Please kindly refer to these new versions in our reply to you. We have attached this new track changes version here for your convenience. Thank you.
  
  Citation: https://doi.org/10.5194/egusphere-2025-2546-AC6

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

AR by Hamed Ibrahim on behalf of the Authors (19 Sep 2025) Author's response Author's tracked changes Manuscript

ED: Reconsider after major revisions (26 Sep 2025) by John M. Huthnance

Dear Authors
Thank-you for your revised manuscript. In view of the substance of some of their comments and the considerable ensuing changes, I intend to send your revised manuscript back to the reviewers in due course. However, I think there are some points to attend to first; please see “Detailed comments” below.
Yours sincerely
John Huthnance (editor)

Detailed comments (Line numbers refer to your revised manuscript unless otherwise stated).
Despite your response to reviewer 1 about “usage of the word “shift””, it still appears far too often – even three times on line 5 and nine times in the abstract – where a more specific word would be appropriate and add meaning. For example (line 5), “steric shifts” is unclear, sea level either rises or falls.
Your response to Referee 1 comment “Line 287-288: Could you provide a justification to why the purple box chosen in the western subpolar north Atlantic is representative?” shows a map FIG C1 with the purple box some way south of the purple box in the manuscripts Fig. 10a.
Line 77. “pressure at the ocean bathymetry (i.e., ocean bottom pressure);” –> “ocean bottom pressure”
Line 145 and equation (4). It seems to me that (4) gives the CHANGE of x during each period.
Line 192. “without the annual cycle not removed” is a triple negative! From the figure I think you mean “retaining the annual cycle”. In figure 3 caption I think “annual mean” –> “mean annual cycle?
Lines 230-234, 239-241. Do you need to repeat all these Table values in the text?
Lines 442-443. “somewhat not well recognized” is awkward. Omit “somewhat”?
Table A2 caption should make clear that the values shown are period-2 mean minus period-1 mean (c.f. text line 228).

Hide

AR by Hamed Ibrahim on behalf of the Authors (27 Sep 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (01 Oct 2025) by John M. Huthnance

RR by Anonymous Referee #2 (09 Oct 2025)

RR by Anonymous Referee #1 (20 Oct 2025)

Suggestions for revision or reasons for rejection

We thank the authors for the revised manuscript. It is appreaciated that the authors tried to take into account my major comments by introducing the additional tasks in the introduction. It is however still not ready for publication in my opinion (cf comments below).
Abstract: In the summary of the results, it is not clear which of the task 1 and 2 is answered (still explaining only the sea level increase not its pausing).
In caption of figure 2, it is said at the beginning ‘annual cycle removed’ and towards the end of the caption it is said a ‘13-month low-pass filter is applied’. It is a bit confusing. How is the annual cycle removed? Is it using the 13-month low pass filter or using another method? Please clarify.
Section 3.1: maybe there is more to say than just the balance in period 1 and the imbalance in period 2. One of the results concluded here is ‘halosteric effects is relatively more important in subdomain A during both periods’. To me this is of secondary importance. A striking result of table A1 is for period 2 the change of sign of the steric component (in the whole domain, and subdomains A and B) compared to period 1 (which ultimately is responsible of the pausing of the sea level increase as the mass component on the other hand increases), which becomes negative because of the thermosteric component which is strongly negative: meaning cooling of the region. To me in this section and from the table A1, the conclusions can go a bit beyong what is said:
to answert task 1: Why sea level is rising during period 1? Mostly because of the steric effect, both thermosteric and halosteric for the whole domain, and mostly halosteric in domain A whereas it is mostly thermosteric in domain B. Why sea level is pausing in P2? Because of the compensation between steric and mass effects with the noticeable change of the sign of the steric component driven by the thermosteric component in both subdomains.
Line 258-259: I think there is no density change associated with the decrease of salinity mentioned here as the dots in the T-S diagram do not cross isopycnals but rather move along them. To better see the change in the vertical structure of density, it it possible to add on panels a1 and b1 the vertical density profile.
Methodological issue: doing the difference of the mean states of two different periods where the mean is not actually representative of the period evolution (because the two periods most of the time display trends for the variables studied here) is not appropriate to answer the question asked ( why is there a transition from rising sea level to pausing sea level?). The evolution of the curves during these two periods are so different that the difference between the mean of these two periods is misleading. The conclusion line 285 “Therefore, halosteric expansion is the dominant driver of the transition from rising state to the hiatus state in subdomain A, while thermosteric expansion is the dominant driver in subdomain B.” is not correct. It does not meake sense: Transition from rising state to hiatus state means the sea level stopped increasing which is not compatible with either a halosteric expansion nor a thermosteric expansion. It would be appropriate if anything to compare the linear trends computed over the 2 periods separately. What the authors are saying is that the sea level is relatively higher on average during P2 than during P1 because of […] This statement is different from what the authors are claiming trying to answer.
The transition rom a rising sea level period (positive sea level trend during P1) to a ‘hiatus’ (fairly stable sea level, slghtly negative sea level trend) is readily visible from the time series, with no need to compute means over the period nor differences: for example, for subdomain A (for which the ‘freshening’ is studied in section 4), we can clearly see from the steric, thermostreric, and halosteric time series, and from table A1, that during period 1 the steric sea level increase is due to an increase in halosteric sea level ( while the thermosteric sea level stays fairly stable). During period 2, the steric sea level decreases because of a thermosteric sea level strong decrease (while the halosteric sea level actually keeps increasing but not enough the counterbalance the strong effect of the thermosteric sea level decrease on the steric sea level). So for subdomain A, the sea level pausing is due to a cooling of the area.
In my opinion, the authors are focusing on something else: the authors chose to focus on the importance of the salinity/freshwater change in the domain (and subdomain A and B) —regardless of the ‘hiatus’—and its effect on sea level change.
Section 4: I disagree with the statement line 331-332.
The cross-correlation: To me the correlation map and the time lag map (fig 10) do not provide evidence of the link between salinity in the subpolar north Atlantic and in subdomain A. For the maps on fig 10 to really show a link, the time lag corresponding to the max of correlation should decrease as the paths get closer to subdomain A which is not the case ( the tima lag is mostly zero for the cyan path, the time lag does not decrease for the green path). The path connecting the two pathways is not evidenced by fig 11a: on figure 11a, where the connexion presumably is, the flow is northeastward and corresponds to the North Atlantic Current. In addition, these hypothesised paths (southward flow for the green path and eastward flow for the cyan path) are not in agreement with the known circulation pattern of that region of the North Atlantic.
The potential vorticity: line 344: please add a refence here. The two density surfaces chosen ( 1025 and 1037 kg m-3) encompass almost the whole water column and thus appear inappropriate for the application here. Line 354-355: I disagree with that statement, the paths are not visually clear on the PV ratio map.
Line 366: I disagree with this statement. The sea level rise hiatus is not caused by freshening in subdomain A but rather mostly because of cooling (freshening on the contrary causes sea level to rise which is not the case as the sea level stopped rising during period 2 compared to period 1). It is true that the halosteric sea level increases during period 2 at approximately the same rate as during period 1 but that should not be the focus of this paper as this does not explain the sea level hiatus. In addition, the freshening of subdomain A is not convingly explained by the advection of water following paths 1 and 2 in my opinion.
Section 5
As for the first version of this manuscript, the focus in this section is on the sea level increase between the two periods which is different from the claimed aim of studying the transition from rising sea level to pausing sea level.
Conclusions:
Its point 1: This conclusion again focuses on the overall long term higher sea level during period 2 than period 1 which overlooks the sea level variations of the time series on wich the idea of this whole study is based on: the different sea level trends over the respective periods.
The conclusion does not comment on the tasks (1, 2, 3) mentionned in the introduction.

Hide

ED: Reconsider after major revisions (03 Nov 2025) by John M. Huthnance

AR by Hamed Ibrahim on behalf of the Authors (20 Feb 2026) Author's response Author's tracked changes Manuscript

ED: Reconsider after major revisions (05 Mar 2026) by John M. Huthnance

Dear Authors
Thank-you for your response to comments and the revised manuscript. It still appears to me that the path from sub-box 15 in the NW Atlantic to sub-domain A is very tenuous. You refer in line 347 to figure C1 to motivate the NW Atlantic as an area of freshening to correlate with sub-domain A and proceed to identify sub-box 15 as having the maximal correlation. However, given the range of correlation values in Table C1, some even negative, it is not clear that this maximal correlation is really statistically significant rather than a chance extreme among the 30 values. Figure C1 also shows freshening near the coast immediately to the north of sub-domain A as more obvious input to sub-domain A. In any case, to seek an arrival path, it would seem more appropriate to look at an equivalent to figure 10a but showing the cross-correlation with sub-domain A rather than sub-box 15. That would avoid the apparent presumption about the source in the present figure 10a and would indicate how far “upstream” (from sub-domain A) the association can be taken. The present figures 10a and 10b have the difficulty of taking the path through decreasing and increasing correlation and time lag. [Another – probably more convincing – approach to the “source” question is to use an adjoint model run backwards in time; this shows the influences on properties at the time and location of interest. However, this is probably too much to expect in view of the extra “machinery” involved.]
I am sorry that this appears to be a referee (rather than editorial) comment but I am seeking to resolve the comment of referee 1 without the delay of having to find another referee. It is owing to the above that I am asking for ”major” revision.
Detailed comments.
Lines 4, 5 (and anywhere else). Better “rising period” –> “period of rise”
Lines 7-8. “thermosteric expansion that counteracted halosteric expansion”. Surely one of these “expansion” (the first c.f. lines 227-229) should be “contraction”.
Line 8. At this stage your particular use of “multidecadal” is not known to the reader.
Figure 4 caption refers to panel (e) (ORAS5 potential temperature) but the panel is missing.
Line 181. “dats” –> “data”.
Line 249. “loses” –> “losses”
Figure 7 caption. I suggest the same methodology for “(annual cycle removed)” be added here as in the Figure 4 caption (assuming it is the same methodology).
Line 339. What is the basis for the currents shown in figures 9, 11? I guess ORAS5 but missed any statement of that.
Lines 346, 364, 365, 457-458. “decreasing trend” means the trend is getting smaller. I think you mean “negative trend” or simply “decrease”.
Line 393. “there” –> “their”
Line 394. “. . to the northwest African coast.”
Line 400. “. . characterizes . .”
Lines 423-426. “mean sea level is higher in subdomain B compared to subdomain A (Fig. 1[b]), and the multidecadal increase in sea level is larger in subdomain B compared to subdomain A (Table B1). There must therefore be a permanent north-south current between the two subdomains to balance the pressure gradient associated with the slope of the free ocean surface.” The logic here is not clear. B is south of A. Higher msl at B would imply surface geostrophic flow to the east and/or friction-retarded flow from south to north.
Lines 467-469. This sentence is unclear; what does “the same phenomena” refer to? Do you perhaps mean “. . these phenomena varying on a multidecadal timescale.”?
Conclusions. I think you should emphasise that item (i) refers to 1996-2004 and item (ii) to 2010-2018, whereas item (iii) refers to the difference between the means of 1996-2004 and 2010-2018. This is to show that you answer the two objectives from the Introduction.

Yours sincerely
John Huthnance (editor)

Hide

AR by Hamed Ibrahim on behalf of the Authors (27 May 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (11 Jun 2026) by John M. Huthnance

Dear Authors
Thank-you for your revised manuscript. Although the source of freshening is not conclusive, I think the description of the possibility of remote input is now much better balanced. Given all the reviewing this manuscript has received, I am now accepting it subject to “Technical corrections” according to the “Detailed comments” below (mainly for grammar and clarity/accuracy). I do not need to see it again; please upload your revised version to the Copernicus/OS editorial system for publication. Please be aware that there will be no copy-editing and that on publication all the comments will become publicly available.
Thank-you for publishing in Ocean Science.
Yours sincerely
John Huthnance (editor)

Detailed comments.
Lines 11-12. You have not defined “northern subdomain” or “southern subdomain”. Either precede them by “a” not “the” or perhaps preferably define them.
Line 68. Better “mass and atmospheric . .”
Figure 3 panels a1, b1, c1. I think “annual mean” –> “mean annual cycle”.
Figure 4 caption line 1. “. . (mean annual cycle removed) . .”?
Line 330. “. . waters . . are” or “. . water . . is”
Line 337. What is the basis for the currents shown in figures 9, 11? I guess ORAS5 but missed any statement of that.
Line 347. “. . lags . . are sensitive . .”
Line 423. “shows” –> “show”
Lines 439. This phrase “convergence . . surface” is an absolute statement but should only be a difference based only the CHANGES in lines 437-438. There might just be less divergence and less upwelling as in line 438. Then line 441 can state that the change is sufficient to make the absolute statement true.
Figure 13 caption line 1. “. . (mean annual cycle removed) . .”?
Lines 470-472. I think this sentence could be simplified. “sea level” appears twice, and you might just refer to the difference between the two means, omitting “multidecadal increase”.
Line 472. Add “subdomain” after “southern”.
Line 473. “expansion” (spelling)
Line 482. Omit last “the”
Line 491. “. . adjustments . . appear . .”
Line 496. I think temperature should be mentioned along with salinity as a low-frequency cause (albeit not necessarily “remote”).

Hide

AR by Hamed Ibrahim on behalf of the Authors (16 Jun 2026) Author's response Manuscript

Report abuse

Please provide a reason why you see this comment as being abusive.
You might include your name and email but you can also stay anonymous.

Please provide a reason why you see this comment as being abusive.

Please confirm reCaptcha.

Comment*

Name:

Email:

Share

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

RC1: 'Comment on egusphere-2025-2546', Anonymous Referee #1, 06 Aug 2025

This manuscript aims at understanding the causes of the multidecadal sea level rise hiatus that occurred in 2010-2019 in the eastern north tropical Atlantic margin off northwest Africa. To do so, it mainly uses the ocean reanalysis ORAS5, performs an empirical orthogonal function analysis to define subdomains, and decomposes the sea level anomaly into its different components.
This manuscript is interesting, of good scientific quality, well structured, and well-written making it easy to follow the stream of thoughts. It addresses the scientific question of better understanding the drivers of sea level variability in the margin off northwest Africa and how it could be linked to remote processes.
I have two major comments and a few minor remarks which I think should be addressed before publication.
Major comment:
My first major comment concerns the scientific question being asked and which is not aligned with the analyses and results. It seems to me that the analyses presented do not answer the question aimed by the authors (specified lines 43-45) which is to find why the sea level stopped rising over period 2. Instead, by comparing the multi-year mean sea level and its components between the two periods, the authors answer a different scientific question which is: Why is the sea level higher during period 2 than during period 1? The concluding remarks section and the results summarized in the abstract indeed highlight the causes of the higher SLA during period 2 compared to period 1: mostly a halosteric expansion in subdomain A and a thermosteric expansion in subdomain B. The mass of the domain is also greater during period 2 than during period 1 mostly because of net precipitation in subdomain B and a net inflow of freshwater in subdomain A.
In my understanding, the analyses that would have answered the question asked by the authors (which is: why the sea level stopped rising during period 2?) would have been to study separately the two periods and their drivers and then compare them. For example, for the whole domain, the positive sea level trend during period one (fig 4a) seems to be due to a positive steric trend (fig 5a) and a positive manometric trend (fig 4b). On the other hand, during period 2, the stabilisation of the SLA (fig 4a) seems to be due to a positive manometric trend (fig 4b) being counterbalanced by a negative steric trend (fig 5a) which is itself due to a negative thermosteric trend (fig 8a). Therefore, we can interpret the pausing in sea level rise during period 2 as a result of a cooling of the domain.
I suggest that the author adjust their narrative so that the results match the scientific question being asked by either modifying the scientific question or the results presented.

My second major comment is about the potential physical linkage between subdomain A salinity and salinity in the western subpolar north atlantic. In my opinion, the manuscript does not provide enough evidence of the existence of the path 1 and path 2 and the westward flow connecting path 2 and 1. It is not clear to me, to what extent the paths 1 and 2 are already known (please provide citation if so) or if they are an original finding of the authors. These paths could be valid but should be presented with more care (as hypotheses or as potential connections) if they are not demonstrated more convincingly. It would be possible to prove a direct link for example by using water parcels lagrangian backwards tracking in time with particles released in the region of interest.

Minor comments:
Throughout the whole manuscript:
Usage of the word ‘upward’. I found a bit confusing the usage of ‘upward’ to refer to the positive shift in sea level between the two periods. The instinctive understanding of ‘upward’ usually refers to something moved higher in depth (it is used as such lines 9, 29, and 30). I recommend replacing it by ‘positive’ when referring to a shift when values are greater during period 2 than period 1.
Usage of word ‘shift’: This word is overly used and sometimes its meaning is not clear. I suggest replacing it with words like ‘change’, ‘increase’, ‘decrease’ etc. when relevant.
The word ‘hiatus’ is used very early in the manuscript but is never really defined. It is only line 45 than the reader guess than it means ‘pause in the sea level rise’. I recommend defining it early in the introduction. In addition, if the authors decide to modify the scientific question of the manuscript following my major comment, I recommend to not use the word ‘hiatus’ at all as it can be easily replaced by ‘pausing in sea level rise’ or other simple words.

All figures showing time series: It would greatly help the reader to highlight period 1 and 2 on all time series either by lightly color-shading the background or delimiting them using vertical lines.

line 5: The steric and mass contributions do not sum up to 100% which can be confusing as the reader usually expects it to do so. The reader later understands why (table A1) but I suggest clarifying this in the abstract.
Line 20-22: please add a citation to support this sentence.
Line 30: ‘Fig 10’. It seems that the figure number is incorrect. Please verify.
Line 35: ‘temperature shifted upward’. Does the author mean ‘northward’? or increase in temperature? Please clarify.
Line 81-84: It is not clear to me how the vertical land displacement is not taken into account in the study. Is it simply neglected? Is it included in the ORAS5 reanalysis or not? How does this impact the comparison with the altimetry data and grace data?

Typos and misspelling:
First paragraph: spaces are missing in units presentation. This would increase readability.
Line 83: tide gauge is misspelled ‘guage’.
Line 162: ‘gauge’ river is misspelled.
Figure 11: ‘and’ is duplicated in the caption.
Line 379: ‘here’ to replace by ‘there’

Methods:
Line 102: it is possible to simplify the sentence by removing ‘taking the difference between two satellite measurements (i.e.,’.
Line 111: replace ‘bathymetry’ by ‘seabed’ or ‘ ocean floor’
Line 153: The agreement is good between ORAS5 G and Grace data for most of the time series but not in 2003-2005 and from 2015. It would be nice to have some comments on this especially because these years are part of periods one and two.
Figure 3: panels a1 and a2 are not referred nor commented on in the text. What is the relevance of these panels?

Results:
Line 202-203: Can you elaborate more to explain why the interpretation is consistent with the SLA pattern in the EOF analysis? This is not evident to me.
Lines 208-220: These paragraphs were a bit difficult to read. I suggest reorganizing them to separate the interpretation for domain A and for domain B.
Figure 7: I am wondering why analysing the temperature and salinity instead of analysing directly the thermosteric and halosteric sea level. Is it because it is not possible to compute the thermosteric and halosteric sea level for the four layers?
Figure 9 and 12: It would help the reader to place the domain, subdomain A, and subdomain B on these maps when they are part of the plot to help to locate them. On figure 9 a1, it would also help to have a schematic arrow to represent the guinea done cyclonic circulation.
Line 273-274: It is said that layers 2 and 3 have ‘comparatively large salinity shifts’. I don’t understand. It is compared to what? To period 1? To the other layers?
Figure 10: panel a. It would be useful to show the time lag corresponding to the maximum correlation shown on panel a. This would strengthen the hypothesis of the potential pathways to link the subpolar North Atlantic and domain A.
Line 281-282: the currents located around 45N that would link path 2 to path 1 is in my opinion not visible from fig 11a.
Figure 10b: I am wondering about the relevance of this map of regression coefficients. Are the regression coefficients computed with a time lag (I understand it is not but please clarify)? I also noticed that it is barely used in the manuscript. To me the paths drawn on top of the map are not hinted at by the map itself. In other words, in my opinion, this map does not add any evidence to the possible existence of the paths indicated by the arrows. In addition, from a methodological point of view, it is not clear to me why the linear regression during period 2 between salinity averaged in domain A and the other grid points in the Atlantic would prove a causal link between the two, especially because (as it is said in the manuscript) it would take some time for the salinity in one place to be reflected in the other.
Line 287-288: Could you provide a justification to why the purple box chosen in the western subpolar north Atlantic is representative?

Line 291 and following: I am not particularly knowledgeable on potential vorticity. Is the conservation of potential vorticity a common way to try to trace water masses? If so please provide citations.
Line 303-305: To me the paths 1 and 2 are not visually clear in figure 11b.
Line 307-308: How is this lagged correlation computed? Is it from the time series of figure 7 a3? In that case it is natural that this is consistent with figure 7 a3.

Section 5: It seems that for the demonstration of this section the vertical velocities are very important but they are not analysed. Why not also analyse the vertical velocities?
Line 361-364: the shift of the northern edge of the countercurrent is not clear to me.Maybe it would be useful to indicate on the figure.
Line 364: ‘vertical velocity’. The shift of the vertical velocities is not clearly demonstrated but inferred. Please correct the wording.
Concluding section:
Point 5: Can you elaborate more on this point? Do you have examples of work that has been done using numerical simulations in that region or elsewhere? Can you hint at what model would be a good option to do this?

Citation: https://doi.org/10.5194/egusphere-2025-2546-RC1

AC1: 'Reply on RC1', Hamed Ibrahim, 19 Sep 2025

Dear Reviewer,
Thank you for reviewing our manuscript and for providing us with very insightful comments. We have responded to all the comments as documented in the attached pdf file.
Regards,
Hamed,

Citation: https://doi.org/10.5194/egusphere-2025-2546-AC1

AC4: 'Reply on AC1', Hamed Ibrahim, 19 Sep 2025

Revised manuscript for your reference.

Citation: https://doi.org/10.5194/egusphere-2025-2546-AC4

AC5: 'Reply on RC1', Hamed Ibrahim, 27 Sep 2025

Dear Referee,
In light of comments from the handling editor. We have submitted a new revised manuscript and a new track changes version. Please kindly refer to these new versions in our reply to you. We have attached this new track changes version here for your convenience. Thank you.

Citation: https://doi.org/10.5194/egusphere-2025-2546-AC5

RC2: 'Comment on egusphere-2025-2546', Anonymous Referee #2, 08 Aug 2025

This study provides a valuable analysis of the 2010–2018 sea level rise hiatus off northwest Africa using satellite/reanalysis data (1993–2018). The identification of steric dominance (74%) and subpolar-tropical salinity linkages is novel and impactful. Methodological rigor and validation with independent datasets are key strengths. However, clarification of physical mechanisms and data limitations is needed for full acceptance. I recommend acceptance after minor revisions.
Eq. 6 oversimplifies vertical-horizontal coupling. Could you show decadal trends of vertical velocity (w) in the dome core using ORAS5 data?

Please discuss how anticyclonic circulation modulates downwelling against background upwelling in more detail.

While mentioned in Section 2.2, error quantification is missing. Could the authors add error bars in Fig. 4?

The 2010–2018 hiatus (Fig. 2a) coincides with subdomain A freshening (Fig. 7a2) but the causal sequencing is unclear. Please add lead-lag correlation analysis between subpolar salinity and SLA.

Citation: https://doi.org/10.5194/egusphere-2025-2546-RC2

AC2: 'Reply on RC2', Hamed Ibrahim, 19 Sep 2025

Dear Reviewer,
Thank you for reviewing our manuscript and for providing insightful comments that have improved our work. We have responded to all the comments as documented in the attached pdf file.
Regards,
Hamed

Citation: https://doi.org/10.5194/egusphere-2025-2546-AC2

AC3: 'Reply on AC2', Hamed Ibrahim, 19 Sep 2025

Revised manuscript for your reference.

Citation: https://doi.org/10.5194/egusphere-2025-2546-AC3

AC6: 'Reply on RC2', Hamed Ibrahim, 27 Sep 2025

Dear Referee,
In light of comments from the handling editor. We have submitted a new revised manuscript and a new track changes version. Please kindly refer to these new versions in our reply to you. We have attached this new track changes version here for your convenience. Thank you.

Citation: https://doi.org/10.5194/egusphere-2025-2546-AC6

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

AR by Hamed Ibrahim on behalf of the Authors (19 Sep 2025) Author's response Author's tracked changes Manuscript

ED: Reconsider after major revisions (26 Sep 2025) by John M. Huthnance

Dear Authors
Thank-you for your revised manuscript. In view of the substance of some of their comments and the considerable ensuing changes, I intend to send your revised manuscript back to the reviewers in due course. However, I think there are some points to attend to first; please see “Detailed comments” below.
Yours sincerely
John Huthnance (editor)

Detailed comments (Line numbers refer to your revised manuscript unless otherwise stated).
Despite your response to reviewer 1 about “usage of the word “shift””, it still appears far too often – even three times on line 5 and nine times in the abstract – where a more specific word would be appropriate and add meaning. For example (line 5), “steric shifts” is unclear, sea level either rises or falls.
Your response to Referee 1 comment “Line 287-288: Could you provide a justification to why the purple box chosen in the western subpolar north Atlantic is representative?” shows a map FIG C1 with the purple box some way south of the purple box in the manuscripts Fig. 10a.
Line 77. “pressure at the ocean bathymetry (i.e., ocean bottom pressure);” –> “ocean bottom pressure”
Line 145 and equation (4). It seems to me that (4) gives the CHANGE of x during each period.
Line 192. “without the annual cycle not removed” is a triple negative! From the figure I think you mean “retaining the annual cycle”. In figure 3 caption I think “annual mean” –> “mean annual cycle?
Lines 230-234, 239-241. Do you need to repeat all these Table values in the text?
Lines 442-443. “somewhat not well recognized” is awkward. Omit “somewhat”?
Table A2 caption should make clear that the values shown are period-2 mean minus period-1 mean (c.f. text line 228).

Hide

AR by Hamed Ibrahim on behalf of the Authors (27 Sep 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (01 Oct 2025) by John M. Huthnance

RR by Anonymous Referee #2 (09 Oct 2025)

Suggestions for revision or reasons for rejection

The authors have addressed all my questions. I have no further reviews.

Hide

RR by Anonymous Referee #1 (20 Oct 2025)

Suggestions for revision or reasons for rejection

We thank the authors for the revised manuscript. It is appreaciated that the authors tried to take into account my major comments by introducing the additional tasks in the introduction. It is however still not ready for publication in my opinion (cf comments below).
Abstract: In the summary of the results, it is not clear which of the task 1 and 2 is answered (still explaining only the sea level increase not its pausing).
In caption of figure 2, it is said at the beginning ‘annual cycle removed’ and towards the end of the caption it is said a ‘13-month low-pass filter is applied’. It is a bit confusing. How is the annual cycle removed? Is it using the 13-month low pass filter or using another method? Please clarify.
Section 3.1: maybe there is more to say than just the balance in period 1 and the imbalance in period 2. One of the results concluded here is ‘halosteric effects is relatively more important in subdomain A during both periods’. To me this is of secondary importance. A striking result of table A1 is for period 2 the change of sign of the steric component (in the whole domain, and subdomains A and B) compared to period 1 (which ultimately is responsible of the pausing of the sea level increase as the mass component on the other hand increases), which becomes negative because of the thermosteric component which is strongly negative: meaning cooling of the region. To me in this section and from the table A1, the conclusions can go a bit beyong what is said:
to answert task 1: Why sea level is rising during period 1? Mostly because of the steric effect, both thermosteric and halosteric for the whole domain, and mostly halosteric in domain A whereas it is mostly thermosteric in domain B. Why sea level is pausing in P2? Because of the compensation between steric and mass effects with the noticeable change of the sign of the steric component driven by the thermosteric component in both subdomains.
Line 258-259: I think there is no density change associated with the decrease of salinity mentioned here as the dots in the T-S diagram do not cross isopycnals but rather move along them. To better see the change in the vertical structure of density, it it possible to add on panels a1 and b1 the vertical density profile.
Methodological issue: doing the difference of the mean states of two different periods where the mean is not actually representative of the period evolution (because the two periods most of the time display trends for the variables studied here) is not appropriate to answer the question asked ( why is there a transition from rising sea level to pausing sea level?). The evolution of the curves during these two periods are so different that the difference between the mean of these two periods is misleading. The conclusion line 285 “Therefore, halosteric expansion is the dominant driver of the transition from rising state to the hiatus state in subdomain A, while thermosteric expansion is the dominant driver in subdomain B.” is not correct. It does not meake sense: Transition from rising state to hiatus state means the sea level stopped increasing which is not compatible with either a halosteric expansion nor a thermosteric expansion. It would be appropriate if anything to compare the linear trends computed over the 2 periods separately. What the authors are saying is that the sea level is relatively higher on average during P2 than during P1 because of […] This statement is different from what the authors are claiming trying to answer.
The transition rom a rising sea level period (positive sea level trend during P1) to a ‘hiatus’ (fairly stable sea level, slghtly negative sea level trend) is readily visible from the time series, with no need to compute means over the period nor differences: for example, for subdomain A (for which the ‘freshening’ is studied in section 4), we can clearly see from the steric, thermostreric, and halosteric time series, and from table A1, that during period 1 the steric sea level increase is due to an increase in halosteric sea level ( while the thermosteric sea level stays fairly stable). During period 2, the steric sea level decreases because of a thermosteric sea level strong decrease (while the halosteric sea level actually keeps increasing but not enough the counterbalance the strong effect of the thermosteric sea level decrease on the steric sea level). So for subdomain A, the sea level pausing is due to a cooling of the area.
In my opinion, the authors are focusing on something else: the authors chose to focus on the importance of the salinity/freshwater change in the domain (and subdomain A and B) —regardless of the ‘hiatus’—and its effect on sea level change.
Section 4: I disagree with the statement line 331-332.
The cross-correlation: To me the correlation map and the time lag map (fig 10) do not provide evidence of the link between salinity in the subpolar north Atlantic and in subdomain A. For the maps on fig 10 to really show a link, the time lag corresponding to the max of correlation should decrease as the paths get closer to subdomain A which is not the case ( the tima lag is mostly zero for the cyan path, the time lag does not decrease for the green path). The path connecting the two pathways is not evidenced by fig 11a: on figure 11a, where the connexion presumably is, the flow is northeastward and corresponds to the North Atlantic Current. In addition, these hypothesised paths (southward flow for the green path and eastward flow for the cyan path) are not in agreement with the known circulation pattern of that region of the North Atlantic.
The potential vorticity: line 344: please add a refence here. The two density surfaces chosen ( 1025 and 1037 kg m-3) encompass almost the whole water column and thus appear inappropriate for the application here. Line 354-355: I disagree with that statement, the paths are not visually clear on the PV ratio map.
Line 366: I disagree with this statement. The sea level rise hiatus is not caused by freshening in subdomain A but rather mostly because of cooling (freshening on the contrary causes sea level to rise which is not the case as the sea level stopped rising during period 2 compared to period 1). It is true that the halosteric sea level increases during period 2 at approximately the same rate as during period 1 but that should not be the focus of this paper as this does not explain the sea level hiatus. In addition, the freshening of subdomain A is not convingly explained by the advection of water following paths 1 and 2 in my opinion.
Section 5
As for the first version of this manuscript, the focus in this section is on the sea level increase between the two periods which is different from the claimed aim of studying the transition from rising sea level to pausing sea level.
Conclusions:
Its point 1: This conclusion again focuses on the overall long term higher sea level during period 2 than period 1 which overlooks the sea level variations of the time series on wich the idea of this whole study is based on: the different sea level trends over the respective periods.
The conclusion does not comment on the tasks (1, 2, 3) mentionned in the introduction.

Hide

ED: Reconsider after major revisions (03 Nov 2025) by John M. Huthnance

Dear Authors
I now have the referees comments on your revised manuscript. One is content but the other is not; I agree with the criticisms except where I say otherwise. Hence I am asking for major revision.
Please see the referee comments below with my own <> and added at the end.
Unfortunately I may then have to find another referee; the critical referee has declined to see it again.
Yours sincerely
John Huthnance (editor)

Referee comments with some <>
We thank the authors for the revised manuscript. It is appreciated that the authors tried to take into account my major comments by introducing the additional tasks in the introduction. It is however still not ready for publication in my opinion (cf comments below).
Abstract: In the summary of the results, it is not clear which of the task 1 and 2 is answered (still explaining only the sea level increase not its pausing).
In caption of figure 2, it is said at the beginning ‘annual cycle removed’ and towards the end of the caption it is said a ‘13-month low-pass filter is applied’. It is a bit confusing. How is the annual cycle removed? Is it using the 13-month low pass filter or using another method? Please clarify. <>
Section 3.1: maybe there is more to say than just the balance in period 1 and the imbalance in period 2. One of the results concluded here is ‘halosteric effects is relatively more important in subdomain A during both periods’. To me this is of secondary importance. A striking result of table A1 is for period 2 the change of sign of the steric component (in the whole domain, and subdomains A and B) compared to period 1 (which ultimately is responsible of the pausing of the sea level increase as the mass component on the other hand increases), which becomes negative because of the thermosteric component which is strongly negative: meaning cooling of the region. <> To me in this section and from the table A1, the conclusions can go a bit beyond what is said:
to answer task 1: Why sea level is rising during period 1? Mostly because of the steric effect, both thermosteric and halosteric for the whole domain, and mostly halosteric in domain A whereas it is mostly thermosteric in domain B. Why sea level is pausing in P2? Because of the compensation between steric and mass effects with the noticeable change of the sign of the steric component driven by the thermosteric component in both subdomains.
Line 258-259: I think there is no density change associated with the decrease of salinity mentioned here as the dots in the T-S diagram do not cross isopycnals but rather move along them. To better see the change in the vertical structure of density, is it possible to add on panels a1 and b1 the vertical density profile. <>
Methodological issue: doing the difference of the mean states of two different periods where the mean is not actually representative of the period evolution (because the two periods most of the time display trends for the variables studied here) is not appropriate to answer the question asked ( why is there a transition from rising sea level to pausing sea level?). The evolution of the curves during these two periods are so different that the difference between the mean of these two periods is misleading. The conclusion line 285 “Therefore, halosteric expansion is the dominant driver of the transition from rising state to the hiatus state in subdomain A, while thermosteric expansion is the dominant driver in subdomain B.” is not correct. It does not make sense: Transition from rising state to hiatus state means the sea level stopped increasing which is not compatible with either a halosteric expansion nor a thermosteric expansion. It would be appropriate if anything to compare the linear trends computed over the 2 periods separately. What the authors are saying is that the sea level is relatively higher on average during P2 than during P1 because of […] This statement is different from what the authors are claiming trying to answer. <>
The transition from a rising sea level period (positive sea level trend during P1) to a ‘hiatus’ (fairly stable sea level, slightly negative sea level trend) is readily visible from the time series, with no need to compute means over the period nor differences: for example, for subdomain A (for which the ‘freshening’ is studied in section 4), we can clearly see from the steric, thermosteric, and halosteric time series, and from table A1, that during period 1 the steric sea level increase is due to an increase in halosteric sea level ( while the thermosteric sea level stays fairly stable). During period 2, the steric sea level decreases because of a thermosteric sea level strong decrease (while the halosteric sea level actually keeps increasing but not enough the counterbalance the strong effect of the thermosteric sea level decrease on the steric sea level). So for subdomain A, the sea level pausing is due to a cooling of the area.
In my opinion, the authors are focusing on something else: the authors chose to focus on the importance of the salinity/freshwater change in the domain (and subdomain A and B) —regardless of the ‘hiatus’—and its effect on sea level change. <>
Section 4: I disagree with the statement line 331-332. <>
The cross-correlation: To me the correlation map and the time lag map (fig 10) do not provide evidence of the link between salinity in the subpolar north Atlantic and in subdomain A. For the maps on fig 10 to really show a link, the time lag corresponding to the max of correlation should decrease as the paths get closer to subdomain A which is not the case ( the time lag is mostly zero for the cyan path, the time lag does not decrease for the green path). The path connecting the two pathways is not evidenced by fig 11a: on figure 11a, where the connexion presumably is, the flow is northeastward and corresponds to the North Atlantic Current. In addition, these hypothesised paths (southward flow for the green path and eastward flow for the cyan path) are not in agreement with the known circulation pattern of that region of the North Atlantic. <>
The potential vorticity: line 344: please add a reference here. The two density surfaces chosen ( 1025 and 1037 kg m-3) encompass almost the whole water column and thus appear inappropriate for the application here. Line 354-355: I disagree with that statement, the paths are not visually clear on the PV ratio map.
Line 366: I disagree with this statement. The sea level rise hiatus is not caused by freshening in subdomain A but rather mostly because of cooling (freshening on the contrary causes sea level to rise which is not the case as the sea level stopped rising during period 2 compared to period 1). It is true that the halosteric sea level increases during period 2 at approximately the same rate as during period 1 but that should not be the focus of this paper as this does not explain the sea level hiatus. <> In addition, the freshening of subdomain A is not convincingly explained by the advection of water following paths 1 and 2 in my opinion.
Section 5
As for the first version of this manuscript, the focus in this section is on the sea level increase between the two periods which is different from the claimed aim of studying the transition from rising sea level to pausing sea level.
<< Again, regarding these comments on section 5 and the Conclusions, maybe you need to clarify the description in lines 47-51 where there seems to be some repetition and unclear distinction between tasks 1 and 2.>>
Conclusions:
Its point 1: This conclusion again focuses on the overall long term higher sea level during period 2 than period 1 which overlooks the sea level variations of the time series on which the idea of this whole study is based on: the different sea level trends over the respective periods.
The conclusion does not comment on the tasks (1, 2, 3) mentioned in the introduction.

Additional Editor comments
Figure 6 panel a1. Spelling – “Temperature”
Line 281. “and” –> “than”
Lines 462-463. I agree. As well as the doubts about the fresher-water transport from the sub-polar North Atlantic in section 4, I found the description of section 5 rather tenuous and would welcome closer argument.

Hide

AR by Hamed Ibrahim on behalf of the Authors (20 Feb 2026) Author's response Author's tracked changes Manuscript

ED: Reconsider after major revisions (05 Mar 2026) by John M. Huthnance

Dear Authors
Thank-you for your response to comments and the revised manuscript. It still appears to me that the path from sub-box 15 in the NW Atlantic to sub-domain A is very tenuous. You refer in line 347 to figure C1 to motivate the NW Atlantic as an area of freshening to correlate with sub-domain A and proceed to identify sub-box 15 as having the maximal correlation. However, given the range of correlation values in Table C1, some even negative, it is not clear that this maximal correlation is really statistically significant rather than a chance extreme among the 30 values. Figure C1 also shows freshening near the coast immediately to the north of sub-domain A as more obvious input to sub-domain A. In any case, to seek an arrival path, it would seem more appropriate to look at an equivalent to figure 10a but showing the cross-correlation with sub-domain A rather than sub-box 15. That would avoid the apparent presumption about the source in the present figure 10a and would indicate how far “upstream” (from sub-domain A) the association can be taken. The present figures 10a and 10b have the difficulty of taking the path through decreasing and increasing correlation and time lag. [Another – probably more convincing – approach to the “source” question is to use an adjoint model run backwards in time; this shows the influences on properties at the time and location of interest. However, this is probably too much to expect in view of the extra “machinery” involved.]
I am sorry that this appears to be a referee (rather than editorial) comment but I am seeking to resolve the comment of referee 1 without the delay of having to find another referee. It is owing to the above that I am asking for ”major” revision.
Detailed comments.
Lines 4, 5 (and anywhere else). Better “rising period” –> “period of rise”
Lines 7-8. “thermosteric expansion that counteracted halosteric expansion”. Surely one of these “expansion” (the first c.f. lines 227-229) should be “contraction”.
Line 8. At this stage your particular use of “multidecadal” is not known to the reader.
Figure 4 caption refers to panel (e) (ORAS5 potential temperature) but the panel is missing.
Line 181. “dats” –> “data”.
Line 249. “loses” –> “losses”
Figure 7 caption. I suggest the same methodology for “(annual cycle removed)” be added here as in the Figure 4 caption (assuming it is the same methodology).
Line 339. What is the basis for the currents shown in figures 9, 11? I guess ORAS5 but missed any statement of that.
Lines 346, 364, 365, 457-458. “decreasing trend” means the trend is getting smaller. I think you mean “negative trend” or simply “decrease”.
Line 393. “there” –> “their”
Line 394. “. . to the northwest African coast.”
Line 400. “. . characterizes . .”
Lines 423-426. “mean sea level is higher in subdomain B compared to subdomain A (Fig. 1[b]), and the multidecadal increase in sea level is larger in subdomain B compared to subdomain A (Table B1). There must therefore be a permanent north-south current between the two subdomains to balance the pressure gradient associated with the slope of the free ocean surface.” The logic here is not clear. B is south of A. Higher msl at B would imply surface geostrophic flow to the east and/or friction-retarded flow from south to north.
Lines 467-469. This sentence is unclear; what does “the same phenomena” refer to? Do you perhaps mean “. . these phenomena varying on a multidecadal timescale.”?
Conclusions. I think you should emphasise that item (i) refers to 1996-2004 and item (ii) to 2010-2018, whereas item (iii) refers to the difference between the means of 1996-2004 and 2010-2018. This is to show that you answer the two objectives from the Introduction.

Yours sincerely
John Huthnance (editor)

Hide

AR by Hamed Ibrahim on behalf of the Authors (27 May 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (11 Jun 2026) by John M. Huthnance

Dear Authors
Thank-you for your revised manuscript. Although the source of freshening is not conclusive, I think the description of the possibility of remote input is now much better balanced. Given all the reviewing this manuscript has received, I am now accepting it subject to “Technical corrections” according to the “Detailed comments” below (mainly for grammar and clarity/accuracy). I do not need to see it again; please upload your revised version to the Copernicus/OS editorial system for publication. Please be aware that there will be no copy-editing and that on publication all the comments will become publicly available.
Thank-you for publishing in Ocean Science.
Yours sincerely
John Huthnance (editor)

Detailed comments.
Lines 11-12. You have not defined “northern subdomain” or “southern subdomain”. Either precede them by “a” not “the” or perhaps preferably define them.
Line 68. Better “mass and atmospheric . .”
Figure 3 panels a1, b1, c1. I think “annual mean” –> “mean annual cycle”.
Figure 4 caption line 1. “. . (mean annual cycle removed) . .”?
Line 330. “. . waters . . are” or “. . water . . is”
Line 337. What is the basis for the currents shown in figures 9, 11? I guess ORAS5 but missed any statement of that.
Line 347. “. . lags . . are sensitive . .”
Line 423. “shows” –> “show”
Lines 439. This phrase “convergence . . surface” is an absolute statement but should only be a difference based only the CHANGES in lines 437-438. There might just be less divergence and less upwelling as in line 438. Then line 441 can state that the change is sufficient to make the absolute statement true.
Figure 13 caption line 1. “. . (mean annual cycle removed) . .”?
Lines 470-472. I think this sentence could be simplified. “sea level” appears twice, and you might just refer to the difference between the two means, omitting “multidecadal increase”.
Line 472. Add “subdomain” after “southern”.
Line 473. “expansion” (spelling)
Line 482. Omit last “the”
Line 491. “. . adjustments . . appear . .”
Line 496. I think temperature should be mentioned along with salinity as a low-frequency cause (albeit not necessarily “remote”).

Hide

AR by Hamed Ibrahim on behalf of the Authors (16 Jun 2026) Author's response Manuscript

Report abuse

Please provide a reason why you see this comment as being abusive.
You might include your name and email but you can also stay anonymous.

Please provide a reason why you see this comment as being abusive.

Please confirm reCaptcha.

Comment*

Name:

Email:

Share

Journal article(s) based on this preprint

01 Jul 2026
A multidecadal sea level rise and its hiatus in the tropical Atlantic margin off northwest Africa
Hamed D. Ibrahim and Yunfang Sun

Ocean Sci., 22, 2059–2081, https://doi.org/10.5194/os-22-2059-2026,https://doi.org/10.5194/os-22-2059-2026, 2026
Short summary

Short summary

This study characterizes a multidecadal sea level rise and its pause in the tropical North Atlantic margin off northwest Africa hosting important marine fisheries. The pause in sea level rise is owing to temperature-driven seawater contraction that counteracted salt-driven expansion and mass accumulation. Currents originating elsewhere freshened the margin with low-salinity water, highlighting a multidecadal linkage between salinity and sea level anomalies in different North Atlantic regions.

Hide

Hamed Ibrahim and Yunfang Sun

Hamed Ibrahim and Yunfang Sun

Viewed

Total article views: 6,470 (including HTML, PDF, and XML)

HTML PDF XML Total BibTeX EndNote

5,406 796 268 6,470 202 271

HTML: 5,406

PDF: 796

XML: 268

Total: 6,470

BibTeX: 202

EndNote: 271

Views and downloads (calculated since 26 Jun 2025)

Cumulative views and downloads (calculated since 26 Jun 2025)

Viewed (geographical distribution)

Total article views: 6,404 (including HTML, PDF, and XML) Thereof 6,404 with geography defined and 0 with unknown origin.

Country # Views %

Total: 0

HTML: 0

PDF: 0

XML: 0

1

1

Latest update: 12 Sep 2026

Hamed Ibrahim

CORRESPONDING AUTHOR

hameddibrahim@gmail.com

https://orcid.org/0000-0002-4720-3476

University of Toronto, Department of Civil & Mineral Engineering, Toronto, ON, Canada

University of Toronto, School of the Environment, Toronto, ON, Canada

University of Toronto, Department of Physics, Toronto, ON, Canada

Yunfang Sun

CORRESPONDING AUTHOR

yunfangsun@gmail.com

https://orcid.org/0000-0001-6656-2581

University of South Florida, College of Marine Science, St Petersburg, FL, USA

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (4311 KB)

Metadata XML

BibTeX

EndNote

Final revised paper

Short summary

This study characterizes the causes of multidecadal sea level shift in the tropical North Atlantic margin off northwest Africa hosting important marine fisheries. The observed sea level rise hiatus is caused by seawater expansion (74%) due to freshening and warming, and by water accumulation (22%) due to changes in currents. The freshening is linked to subpolar North Atlantic waters, highlighting a multidecadal linkage between salinity and sea level anomalies in different North Atlantic regions.

This study characterizes the causes of multidecadal sea level shift in the tropical North...

Read more

Share