Externally-forced and intrinsic variability of the Mediterranean surface and overturning circulations

Héron, Damien; Penduff, Thierry; Brankart, Jean-Michel; Brasseur, Pierre; Somot, Samuel; Waldman, Robin; Pennel, Romain

doi:10.5194/egusphere-2025-5227

Preprints

https://doi.org/10.5194/egusphere-2025-5227

Preprints

03 Nov 2025

| 03 Nov 2025

Externally-forced and intrinsic variability of the Mediterranean surface and overturning circulations

Damien Héron, Thierry Penduff, Jean-Michel Brankart, Pierre Brasseur, Samuel Somot, Robin Waldman, and Romain Pennel

Abstract. Part of Mediterranean Sea variability is forced and paced by external drivers (e.g. atmosphere, river runoffs, Atlantic inflow); the other part has a random phase and spontaneously emerges due to chaotic intrinsic variability (CIV). This study quantifies across time scales the imprints of both variability components on the surface and zonal overturning circulations within the basin, from a 39-year 30-member ensemble ocean 1/12° simulation. We find that most of SSH variance is intrinsic over 17 % of the basin, in particular in the southern Ionian and Levantine basins, and most notably in the Algerian basin where CIV explains 80 % of the SSH variance at periods greater than 4 months. In contrast, 75 % of its interannual to decadal variance of the North Ionian Gyre circulation is paced by the atmosphere, suggesting an external triggering of the Adriatic-Ionian Bimodal Oscillating System (BiOS) reversal. Other gyres such as Rhodes, Bonifacio, and Alboran shows more balanced contribution of CIV. Fluctuations of the density-coordinate zonal overturning circulation (ZOC_σ) and of associated transports are mostly paced by the forcing over most of the basin. However, CIV tends to explain a larger fraction of these transports variance in the intermediate and bottom layers near deep convection sites, in particular in the Levantine basin where this fraction exceeds 50 % between 27 and 30° E at submonthly periods, and 20–30 % at periods reaching 20 years locally. This partly random character of the multi-scale Mediterranean variability has consequences for evaluating model simulations, and the design of observation systems targeting the long-term monitoring of the basin.

Received: 22 Oct 2025 – Discussion started: 03 Nov 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, 8146 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 (8146 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

11 Feb 2026

Externally-forced and intrinsic variability of the Mediterranean surface and overturning circulations

Damien Héron, Thierry Penduff, Jean-Michel Brankart, Pierre Brasseur, Samuel Somot, Robin Waldman, and Romain Pennel

Ocean Sci., 22, 531–547, https://doi.org/10.5194/os-22-531-2026,https://doi.org/10.5194/os-22-531-2026, 2026

Short summary

Damien Héron, Thierry Penduff, Jean-Michel Brankart, Pierre Brasseur, Samuel Somot, Robin Waldman, and Romain Pennel

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-5227', Anonymous Referee #1, 15 Nov 2025
Héron et al. analyse a 39-year, 30-member, 1/12° ensemble simulations of the Mediterranean Sea to distinguish intrinsic from forced variability in surface and zonal overturning circulation. Ensemble statistics and temporal scale decomposition reveal that SSH variability is predominantly intrinsic over about 17% of the basin at both timescales, with hotspots in the Algerian, Levantine, and Ionian Seas. Zonal overturning variability is largely atmosphere-driven, though intrinsic processes remain important in intermediate and deep layers near convection sites, particularly in the Levantine basin. Using relative vorticity, the authors further show that Mediterranean gyres span a continuum from strongly forced (e.g., North Ionian Gyre) to largely intrinsic (e.g., Algerian Gyre).
The manuscript is clearly written, with some minor structural aspects that could be refined, and supported by robust results. It addresses a relevant scientific question within the broader context of detection and attribution. Although the study is primarily descriptive, it targets a region where such characterization is still limited and provides insights that can meaningfully advance our understanding of Mediterranean Sea dynamics.
I recommend acceptance with the following minor revisions:
I would suggest considering a revision of the abstract, as it currently reads a bit convoluted and may not clearly convey the main results of the study.

L74 - 83: The manuscript provides an extensive description of the ALDERA3 dataset. As I understand, this dataset was not produced by the authors but obtained from another source. I recommend clarifying this point explicitly. In that case, the description could be shortened to include only the essential aspects relevant to the study, with a reference to the original documentation for further details.

Subsection 2.3.3: Could you add a comment on why you chose 2.5 years as a threshold to distinguish the LF and HF components of the timeseries?

L124: If I understood your equation correctly, I think it should be H(σ₀(x,y,z,t,n) - σ₀).

L179 - 185: I would comment differently on the HF SSH variability. The model is in good agreement with the observation and it is indeed underestimating the variability along the Algerian coast and the Levantine basin. However, this is not true in the north-western part of the basin, in the Northern Adriatic Sea, and in the Gulf of Gabes.

Figure 1: I think that the discussion about the specific gyres would be more easily understandable if the boxes shown in Figure 1 were named or referred to in some way in the Figure itself or in the caption. You could use numbers, letters, or acronyms to refer to each specific box. Moreover, I would recommend changing the color of the axes of the left plots to something other than yellow and that of the boxes in panels e and f to something brighter.

In subsection 3.2.2, I would add that the bottom zonal transport at low frequency is comparable to the surface and intermediate zonal transports in the regions 17 - 21°E and 27 - 30°E, and comment on that.

The horizontal lines in panels c and f of Figure 4 are defined later in the text, in section 4.1.3. I would suggest adding that to the Figure’s caption as well.

L275 - 276: Maybe add a short comment on why large values of forced HF SSH variability are found in the Alboran Sea and in the Tunisian shelf.

I would unify subsections 4.1.1 and 4.1.2 into a single one since a comparison between HF and LF variability is made, focusing on the similarities and the differences between the two temporal scales. Could you give an interpretation of the reason why the two R_i maps generally agree except for the Tyrrhenian Sea and the Ionian Sea?

I would suggest doing the same for subsections 4.3.1 and 4.3.2. Could you also further comment on the different results obtained for HF and LF variance of zonal transport, especially about the difference in magnitude between the two?

I would recommend stressing more the relevance of the analysis of individual gyres and the motivation behind it.

Have you checked how dependent your results are on the ensemble size? Please comment on that.

In the conclusion, among the possible deficiencies of the model used, I would emphasize the well-known dependence of internal variability on horizontal resolution. The model has a resolution of 1/12°, which, in the Mediterranean Sea, is often much larger than the Rossby radius of deformation.

Even though you have highlighted the differences in the conducted analysis compared to the previous studies that you cited (Benincasa et al., 2024, Waldman et al., 2018, and the OCCIPUT project), it would be valuable to include a concise but explicit comparison of your findings with the results reported in those works.

Technical corrections:
L6: I suggest writing “.. 75% of the interannual to decadal SSH variance of the North Ionian Gyre circulation ..”.

L12: I suggest rephrasing “and 20-30% at periods reaching 20 years locally” to make it clearer.

The multiplication symbol in the units of several plotted quantities and throughout the text is not rendered well in the document. I would suggest double-checking the LaTeX expression.

L76: Missing parentheses for references “Nabat et al., 2020” and “Colin et al., 2010” on the next line.

L253 - 259: It seems like the same concept, i.e. that the total HF variability of the intermediate westward transport is weaker than that of the eastward surface transport west of 10°E, is repeated two times.

L270 - 271: I would position the introductory sentence “The left .. at LF (bottom)” in section 4.1 before the beginning of subsection 4.1.1.

L290: typo in “Figure4d”.

Figure 5: Specify that you are showing the log of the Power Spectral Density in the caption.

Figure 6: Increase the size of the title of panel b.

L339: typo $R_i(t)$.

L366: typo “behaviors”.

Regarding the multi-scale character of the Mediterranean Sea, I would suggest adding the following reference: Robinson, A. R., Leslie, W. G., Theocharis, A., and Lascaratos, A.: Mediterranean sea circulation, Ocean currents, 2001, 1689– 1705, https://doi.org/10.1006/rwos.2001.0376, 2001.
Citation: https://doi.org/10.5194/egusphere-2025-5227-RC1
- AC2: 'Reply on RC1', Damien Héron, 16 Jan 2026
  
  Publisher’s note: the supplement to this comment was edited on 27 January 2026. The adjustments were minor without effect on the scientific meaning.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5227-AC2
RC2:
'Comment on egusphere-2025-5227', Anonymous Referee #2, 19 Nov 2025
The Authors use a 39-year, 30-member ensemble ocean simulation to assess basin-wide contributions of externally forced and intrinsic variability to key components of Mediterranean circulation across different scales.
Overall, the manuscript is well written, presents a relevant scientific topic, and employs appropriate methods. It is therefore suitable for acceptance after a few minor revisions suggested below.
L33. Can you add one or more references where “chaotic intrinsic variability” (CIV) is first introduced/studied?

L70. I suggest adding the approximate horizontal resolution in kilometers after „1/12°“.

L209. Please check the specified locations of the maximums and the corresponding potential densities in Figure 2a. The listed densities and the second location (24° E) do not appear to match the plotted values.

L213 and L215. Typo „Straits“ to „Strait“.

L355. I recommend rephrasing as: „Contrary to previous Mediterranean Sea multi-model evaluation studies (e.g., Dunić et al., 2019),...“

L357-360 and L428-431. Could you elaborate further on the conclusion that the atmosphere is the predominant external driver of decadal fluctuations of the NIG (i.e., BiOS)?

L434 and L451. Typo „sea“ to „Sea“.

I suggest adding a short discussion on how horizontal resolution may influence the intrinsic variability.
Citation: https://doi.org/10.5194/egusphere-2025-5227-RC2
- AC1: 'Reply on RC2', Damien Héron, 16 Jan 2026
  
  Publisher’s note: the supplement to this comment was edited on 27 January 2026. The adjustments were minor without effect on the scientific meaning.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5227-AC1
EC1: 'Comment on egusphere-2025-5227', Bernadette Sloyan, 02 Dec 2025

The referees have provided several comments and suggestions on the manuscript. Referee 1, In particular, has made a number of suggestions and clarification for the authors to consider. I encourage the authors to provide a point-by-point response to referee comments and revised manuscript to Ocean Sciences.

Citation: https://doi.org/10.5194/egusphere-2025-5227-EC1

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-5227', Anonymous Referee #1, 15 Nov 2025
Héron et al. analyse a 39-year, 30-member, 1/12° ensemble simulations of the Mediterranean Sea to distinguish intrinsic from forced variability in surface and zonal overturning circulation. Ensemble statistics and temporal scale decomposition reveal that SSH variability is predominantly intrinsic over about 17% of the basin at both timescales, with hotspots in the Algerian, Levantine, and Ionian Seas. Zonal overturning variability is largely atmosphere-driven, though intrinsic processes remain important in intermediate and deep layers near convection sites, particularly in the Levantine basin. Using relative vorticity, the authors further show that Mediterranean gyres span a continuum from strongly forced (e.g., North Ionian Gyre) to largely intrinsic (e.g., Algerian Gyre).
The manuscript is clearly written, with some minor structural aspects that could be refined, and supported by robust results. It addresses a relevant scientific question within the broader context of detection and attribution. Although the study is primarily descriptive, it targets a region where such characterization is still limited and provides insights that can meaningfully advance our understanding of Mediterranean Sea dynamics.
I recommend acceptance with the following minor revisions:
I would suggest considering a revision of the abstract, as it currently reads a bit convoluted and may not clearly convey the main results of the study.

L74 - 83: The manuscript provides an extensive description of the ALDERA3 dataset. As I understand, this dataset was not produced by the authors but obtained from another source. I recommend clarifying this point explicitly. In that case, the description could be shortened to include only the essential aspects relevant to the study, with a reference to the original documentation for further details.

Subsection 2.3.3: Could you add a comment on why you chose 2.5 years as a threshold to distinguish the LF and HF components of the timeseries?

L124: If I understood your equation correctly, I think it should be H(σ₀(x,y,z,t,n) - σ₀).

L179 - 185: I would comment differently on the HF SSH variability. The model is in good agreement with the observation and it is indeed underestimating the variability along the Algerian coast and the Levantine basin. However, this is not true in the north-western part of the basin, in the Northern Adriatic Sea, and in the Gulf of Gabes.

Figure 1: I think that the discussion about the specific gyres would be more easily understandable if the boxes shown in Figure 1 were named or referred to in some way in the Figure itself or in the caption. You could use numbers, letters, or acronyms to refer to each specific box. Moreover, I would recommend changing the color of the axes of the left plots to something other than yellow and that of the boxes in panels e and f to something brighter.

In subsection 3.2.2, I would add that the bottom zonal transport at low frequency is comparable to the surface and intermediate zonal transports in the regions 17 - 21°E and 27 - 30°E, and comment on that.

The horizontal lines in panels c and f of Figure 4 are defined later in the text, in section 4.1.3. I would suggest adding that to the Figure’s caption as well.

L275 - 276: Maybe add a short comment on why large values of forced HF SSH variability are found in the Alboran Sea and in the Tunisian shelf.

I would unify subsections 4.1.1 and 4.1.2 into a single one since a comparison between HF and LF variability is made, focusing on the similarities and the differences between the two temporal scales. Could you give an interpretation of the reason why the two R_i maps generally agree except for the Tyrrhenian Sea and the Ionian Sea?

I would suggest doing the same for subsections 4.3.1 and 4.3.2. Could you also further comment on the different results obtained for HF and LF variance of zonal transport, especially about the difference in magnitude between the two?

I would recommend stressing more the relevance of the analysis of individual gyres and the motivation behind it.

Have you checked how dependent your results are on the ensemble size? Please comment on that.

In the conclusion, among the possible deficiencies of the model used, I would emphasize the well-known dependence of internal variability on horizontal resolution. The model has a resolution of 1/12°, which, in the Mediterranean Sea, is often much larger than the Rossby radius of deformation.

Even though you have highlighted the differences in the conducted analysis compared to the previous studies that you cited (Benincasa et al., 2024, Waldman et al., 2018, and the OCCIPUT project), it would be valuable to include a concise but explicit comparison of your findings with the results reported in those works.

Technical corrections:
L6: I suggest writing “.. 75% of the interannual to decadal SSH variance of the North Ionian Gyre circulation ..”.

L12: I suggest rephrasing “and 20-30% at periods reaching 20 years locally” to make it clearer.

The multiplication symbol in the units of several plotted quantities and throughout the text is not rendered well in the document. I would suggest double-checking the LaTeX expression.

L76: Missing parentheses for references “Nabat et al., 2020” and “Colin et al., 2010” on the next line.

L253 - 259: It seems like the same concept, i.e. that the total HF variability of the intermediate westward transport is weaker than that of the eastward surface transport west of 10°E, is repeated two times.

L270 - 271: I would position the introductory sentence “The left .. at LF (bottom)” in section 4.1 before the beginning of subsection 4.1.1.

L290: typo in “Figure4d”.

Figure 5: Specify that you are showing the log of the Power Spectral Density in the caption.

Figure 6: Increase the size of the title of panel b.

L339: typo $R_i(t)$.

L366: typo “behaviors”.

Regarding the multi-scale character of the Mediterranean Sea, I would suggest adding the following reference: Robinson, A. R., Leslie, W. G., Theocharis, A., and Lascaratos, A.: Mediterranean sea circulation, Ocean currents, 2001, 1689– 1705, https://doi.org/10.1006/rwos.2001.0376, 2001.
Citation: https://doi.org/10.5194/egusphere-2025-5227-RC1
- AC2: 'Reply on RC1', Damien Héron, 16 Jan 2026
  
  Publisher’s note: the supplement to this comment was edited on 27 January 2026. The adjustments were minor without effect on the scientific meaning.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5227-AC2
RC2:
'Comment on egusphere-2025-5227', Anonymous Referee #2, 19 Nov 2025
The Authors use a 39-year, 30-member ensemble ocean simulation to assess basin-wide contributions of externally forced and intrinsic variability to key components of Mediterranean circulation across different scales.
Overall, the manuscript is well written, presents a relevant scientific topic, and employs appropriate methods. It is therefore suitable for acceptance after a few minor revisions suggested below.
L33. Can you add one or more references where “chaotic intrinsic variability” (CIV) is first introduced/studied?

L70. I suggest adding the approximate horizontal resolution in kilometers after „1/12°“.

L209. Please check the specified locations of the maximums and the corresponding potential densities in Figure 2a. The listed densities and the second location (24° E) do not appear to match the plotted values.

L213 and L215. Typo „Straits“ to „Strait“.

L355. I recommend rephrasing as: „Contrary to previous Mediterranean Sea multi-model evaluation studies (e.g., Dunić et al., 2019),...“

L357-360 and L428-431. Could you elaborate further on the conclusion that the atmosphere is the predominant external driver of decadal fluctuations of the NIG (i.e., BiOS)?

L434 and L451. Typo „sea“ to „Sea“.

I suggest adding a short discussion on how horizontal resolution may influence the intrinsic variability.
Citation: https://doi.org/10.5194/egusphere-2025-5227-RC2
- AC1: 'Reply on RC2', Damien Héron, 16 Jan 2026
  
  Publisher’s note: the supplement to this comment was edited on 27 January 2026. The adjustments were minor without effect on the scientific meaning.
  
  Citation: https://doi.org/10.5194/egusphere-2025-5227-AC1
EC1: 'Comment on egusphere-2025-5227', Bernadette Sloyan, 02 Dec 2025

The referees have provided several comments and suggestions on the manuscript. Referee 1, In particular, has made a number of suggestions and clarification for the authors to consider. I encourage the authors to provide a point-by-point response to referee comments and revised manuscript to Ocean Sciences.

Citation: https://doi.org/10.5194/egusphere-2025-5227-EC1

Peer review completion

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

AR by Damien Héron on behalf of the Authors (26 Jan 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (05 Feb 2026) by Bernadette Sloyan

AR by Damien Héron on behalf of the Authors (05 Feb 2026) Manuscript

Journal article(s) based on this preprint

11 Feb 2026

Externally-forced and intrinsic variability of the Mediterranean surface and overturning circulations

Damien Héron, Thierry Penduff, Jean-Michel Brankart, Pierre Brasseur, Samuel Somot, Robin Waldman, and Romain Pennel

Ocean Sci., 22, 531–547, https://doi.org/10.5194/os-22-531-2026,https://doi.org/10.5194/os-22-531-2026, 2026

Short summary

Damien Héron, Thierry Penduff, Jean-Michel Brankart, Pierre Brasseur, Samuel Somot, Robin Waldman, and Romain Pennel

Viewed

Total article views: 460 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
271	152	37	460	20	19

HTML: 271
PDF: 152
XML: 37
Total: 460
BibTeX: 20
EndNote: 19

Views and downloads (calculated since 03 Nov 2025)

Month	HTML	PDF	XML	Total
Nov 2025	118	29	15	162
Dec 2025	53	60	11	124
Jan 2026	79	55	11	145
Feb 2026	21	8	0	29

Cumulative views and downloads (calculated since 03 Nov 2025)

Month	HTML	PDF	XML	Total
Nov 2025	118	29	15	162
Dec 2025	53	60	11	124
Jan 2026	79	55	11	145
Feb 2026	21	8	0	29

Viewed (geographical distribution)

Total article views: 440 (including HTML, PDF, and XML) Thereof 440 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 11 Feb 2026

Download

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

Preprint (8146 KB)
Metadata XML

Short summary

Our study used realistic ocean simulations to determine how much of the Mediterranean’s circulation is due to natural randomness rather than atmospheric forcing. We found that spontaneous ocean variability is strong in several regions and can persist for years or even decades. This randomness influences how well models and observations can capture the Mediterranean’s response to climate change.


Total:	0
HTML:	0
PDF:	0
XML:	0