Eddy Kinetic Energy Variability From 30 Years of Altimetry in the Mediterranean Sea

Hargous, Paul; Combes, Vincent; Barceló-Llull, Bàrbara; Pascual, Ananda

doi:10.5194/egusphere-2025-4651

Preprints

https://doi.org/10.5194/egusphere-2025-4651

Preprints

06 Oct 2025

| 06 Oct 2025

Eddy Kinetic Energy Variability From 30 Years of Altimetry in the Mediterranean Sea

Paul Hargous, Vincent Combes, Bàrbara Barceló-Llull, and Ananda Pascual

Abstract. Mesoscale activity plays a central role in ocean variability, substantially influencing the mixing of biogeophysical tracers, such as heat and carbon, and driving changes in ecosystems. Eddy Kinetic Energy (EKE), a metric used for studying the intensity of mesoscale processes, has recently been shown to increase in regions of intense EKE worldwide. Strong EKE positive trends are observed, for example, in the principal western boundary current regions, such as the Gulf Stream, Kuroshio Extension, and the Brazil/Malvinas Confluence. In this study, we assess whether the Mediterranean Sea, known to be a hotspot for climate change impacts, also exhibits such intensification. Despite the high number of observational data and modeling experiments, there is a gap in understanding the long-term evolution of mesoscale dynamics and EKE trends in the Mediterranean Sea. This study investigates EKE trends in the Mediterranean Sea using daily geostrophic currents derived from satellite altimetric data. To test the robustness of the results, we compare EKE trends computed from three different gridded altimetric products: a global product derived from a stable two-satellite constellation (two-sat) and two other products (global and European) incorporating all available satellites (all-sat). While all products reveal a general increase in EKE in the Mediterranean Sea over the last three decades, the trends calculated from the two-sat product are significantly smaller than those computed from the all-sat products. We show that this discrepancy is strongly linked to the increasing number of satellites over time used to construct the all-sat data sets, which enhances both spatial and temporal coverage, and hence, their capacity to detect higher energy levels. To evaluate the fidelity of these gridded products in capturing EKE trends, we compare them with along-track data in high-energy regions of the Mediterranean Sea: the Alboran Sea and the Ierapetra area. These regions exhibit contrasting EKE trends: positive in the Alboran Sea and negative in the Ierapetra area. These findings highlight the importance of using altimetric products with a stable number of satellites constructed for climate applications when addressing long-term ocean variability analysis.

Received: 22 Sep 2025 – Discussion started: 06 Oct 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, 14204 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 (14204 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

12 Feb 2026

Eddy kinetic energy variability from 30 years of altimetry in the Mediterranean Sea

Paul Hargous, Vincent Combes, Bàrbara Barceló-Llull, and Ananda Pascual

Ocean Sci., 22, 549–564, https://doi.org/10.5194/os-22-549-2026,https://doi.org/10.5194/os-22-549-2026, 2026

Short summary

Paul Hargous, Vincent Combes, Bàrbara Barceló-Llull, and Ananda Pascual

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-4651', Anonymous Referee #1, 21 Oct 2025
Eddy Kinetic Energy (EKE) is a widely used metric for ocean currents variability, enabling to monitor them using remote-sensing, particularly through altimetry since 1993. Some earlier studies reported increasing trends in EKE worldwide, which was not yet extensively studied in the Mediterranean Sea. The authors explored the sensibility to the increasing number of altimetric satellites. Section 3.2 explores the differences between all-sat and two-sat Level 4 products ; section 3.3 asserts the results by comparison with along-tracks Level 3 data ; section 4 provides an insight of the statistics using the META eddy atlas.
The method provides an insight to altimetry products biases, often curtailed in many mesoscale studies. The results are quite interesting as they show the number of satellites drives a strong bias on observed long-term trends, in EKE but also potentially on individual eddy detections. They showed that while an EKE trend might appear over the Mediterranean Sea in all-sat product, this is actually not significant using two-sat timeseries. Figure 5 in particular is very interesting - and might be further highlighted - as a mapped sensitivity of satellite sampling.
Section 3,4 using META eddy detections is interesting as it asserts the previous results with eddies statistics. It seems however to stay under-used, as it introduces a totally different dataset for 15 lines of development. I recommend extending this section to illustrates other regions and stregnthen the robustness of the statistics, and also make distinction between cyclones and anticyclone (not possible using EKE only).
A major concern arises about the regions of application, as the study focuses on 2 limited energetic regions to assess its results. These areas (Alboran & Ierapetra) are known to be energetic because of the presence of a single (or two) recurrent mesoscale structures. In the particular case of Ierapetra the reported results seem to be a shift of the structure.
Major comment :
Adding an application of the method to larger areas where mesoscale structures are more numerous and behavior more chaotic should be more significant. Such areas in the Mediterranean Sea could be for instance the Algerian Basin or the Central Ionian Jet. Also increase the area of the Ierapetra focus, for instance by a Levantine Basin-wide focus.

Additional comments :
l.29 ‘A widely used metric […] is the EKE’ : Here it would be helpful to make reference to previous well-known studies to better introduce EKE. For example, Wilkin & Morrow (JGR,1994), Stammer (JPO,1997) or Ferrari & Wunsch (2009).

l.40-50 (but also l.309) : when introducing the Mediterranean Sea, remind that the Rossby deformation radius is much smaller in this area, ranging between 20 and 10km. This area is not well described in Chelton et al (1998), but one can also refer to Kurkin et al (https://doi.org/10.2205/2020ES000737, 2020)

Fig.1 : add isolines of EKE to better reflect gradients

l.140 : ‘A focus on the Eastern Alboran gyre’ : not shown

l.147 : For the Ierapetra region : Can you add a description of your region boundary ? It sounds quite small, and therefore may be quite sensitive to shift in boundaries. A more objective box might be drawn taking an EKE-isoline as boundary

l.156 (and after Table.1, l.193) : The ‘area-weighted mean’ was never defined

l.164 : can you add this fitted polynomial to the plot ? If Correlation coeff is so high (0,85), why not using a linear fit instead, as number of satellite is already roughly linear ?

l.171 : The case of the Alboran sea (high EKE but high sensibility) can be a counter-example. Isn’t it rather almost land-locked areas that are more sensitive ? (Alboran, Adriatic, Levantine basin?)

l.188 : about reversal of Ionian Sea circulation, reference about BIOS : Gacic et al (2010)

Table1 : Not clear what are defined as ‘Mean of all trends’ and ‘Trend of the mean’.

l.234 : ‘ability to detect smaller scales features’ This can be asserted by references such as Amores et al (2018) or Stegner et al (2021).

Section3.4 : as described earlier, this subsection is quite interesting but deserves to be investigated further and applied to other regions. A distinction between cyclone and anticyclone should also be made, at least in figure 8. Actually, it sounds surprising that region with cyclonic activity (e.g. South-West Cretean gyre, Rhodes gyre) do not exhibit more sensitivity to increasing number of satellite (in your Fig.5), as they tend be smaller (see observing system experiment in Stegner et al 2021). Do you think EKE trends are rather driven by anticyclone increasing rotating speed over time, as it might be the case for the Alboran gyres ?

Section 4.4 : you can add some discussion also about decennal cycles in the Ionian Sea, where reversal of the regional circulation are expected to have some effects on the EKE of the Midde Mediterranean Jet.

l.275-276 : This behavior could be checked using the META atlas
Citation: https://doi.org/10.5194/egusphere-2025-4651-RC1
- AC1: 'Reply on RC1', Paul Hargous, 19 Dec 2025
  
  We would like to thank the reviewer for devoting their time to revising our manuscript and for their valuable feedback. Please find our comment by comment response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4651-AC1
RC2:
'Comment on egusphere-2025-4651', Pierre-Marie Poulain, 28 Nov 2025

Satellite altimetry products from 1993 to 2023 are assessed to quantify the temporal evolution of surface geostrophic current variability in the Mediterranean Sea. In particular, trends in Eddy Kinetic Energy (EKE) are estimated using three gridded altimetric products. Along-track datasets and an eddy atlas are also employed for comparison in selected regions. Significant positive and negative EKE trends are identified in the most energetic areas of the basin. The influence of the increasing number of satellites on the observed EKE rise is evaluated. Products based on a constant two-satellite configuration appear adequate for estimating long-term trends, but they fail to capture important dynamical structures in some regions. Therefore, the use of constant three-satellite altimetry products is recommended for future investigations of EKE variability in the Mediterranean Sea.
The paper is well structured, and the English is generally fluent and clear. The figures effectively support the written statements. I recommend the publication of this manuscript in Ocean Science, after the authors address the following specific comments (minor revision).
Specific comments:
L43 and L45. Change “waters” to “water”.
L45. Remove “and freshwater”.
L47. Add the following reference before Escudier et al. : Poulain, P. M., Menna, M., & Mauri, E. (2012). Surface geostrophic circulation of the Mediterranean Sea derived from drifter and satellite altimeter data. Journal of Physical Oceanography, 42(6), 973-990. Add this reference in the reference list.
L76. Provide mode details on the Eddy Atlas. Is it based on ADT or SLA structures?
Caption of Fig. 3., L156 and L193. Change “Area-weighted mean” to “Area-averaged”.
Caption of Fig. 6. Change to “Gray-hatched”.
Table 1. Define “Trend of the mean”.
Caption of Fig. 7. Add info about the L3-ref and all-sat-glo curves.
L216. Change “horizontal velocities and their variability are” to “geostrophic velocity variability is”.
L274. Change “However, some go south-east” to “However, some move south-east”.
L306. Change “applications as the two-sat product” to “applications such as the two-sat product”.

Citation: https://doi.org/10.5194/egusphere-2025-4651-RC2
- AC2: 'Reply on RC2', Paul Hargous, 19 Dec 2025
  
  We would like to thank the reviewer for devoting their time to revising our manuscript and for their valuable feedback. Please find our comment by comment response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4651-AC2
RC3:
'Comment on egusphere-2025-4651', Anonymous Referee #3, 09 Dec 2025

The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4651/egusphere-2025-4651-RC3-supplement.pdf

Citation: https://doi.org/10.5194/egusphere-2025-4651-RC3
- AC3: 'Reply on RC3', Paul Hargous, 19 Dec 2025
  
  We would like to thank the reviewer for devoting their time to revising our manuscript and for their valuable feedback. Please find our comment by comment response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4651-AC3

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-4651', Anonymous Referee #1, 21 Oct 2025
Eddy Kinetic Energy (EKE) is a widely used metric for ocean currents variability, enabling to monitor them using remote-sensing, particularly through altimetry since 1993. Some earlier studies reported increasing trends in EKE worldwide, which was not yet extensively studied in the Mediterranean Sea. The authors explored the sensibility to the increasing number of altimetric satellites. Section 3.2 explores the differences between all-sat and two-sat Level 4 products ; section 3.3 asserts the results by comparison with along-tracks Level 3 data ; section 4 provides an insight of the statistics using the META eddy atlas.
The method provides an insight to altimetry products biases, often curtailed in many mesoscale studies. The results are quite interesting as they show the number of satellites drives a strong bias on observed long-term trends, in EKE but also potentially on individual eddy detections. They showed that while an EKE trend might appear over the Mediterranean Sea in all-sat product, this is actually not significant using two-sat timeseries. Figure 5 in particular is very interesting - and might be further highlighted - as a mapped sensitivity of satellite sampling.
Section 3,4 using META eddy detections is interesting as it asserts the previous results with eddies statistics. It seems however to stay under-used, as it introduces a totally different dataset for 15 lines of development. I recommend extending this section to illustrates other regions and stregnthen the robustness of the statistics, and also make distinction between cyclones and anticyclone (not possible using EKE only).
A major concern arises about the regions of application, as the study focuses on 2 limited energetic regions to assess its results. These areas (Alboran & Ierapetra) are known to be energetic because of the presence of a single (or two) recurrent mesoscale structures. In the particular case of Ierapetra the reported results seem to be a shift of the structure.
Major comment :
Adding an application of the method to larger areas where mesoscale structures are more numerous and behavior more chaotic should be more significant. Such areas in the Mediterranean Sea could be for instance the Algerian Basin or the Central Ionian Jet. Also increase the area of the Ierapetra focus, for instance by a Levantine Basin-wide focus.

Additional comments :
l.29 ‘A widely used metric […] is the EKE’ : Here it would be helpful to make reference to previous well-known studies to better introduce EKE. For example, Wilkin & Morrow (JGR,1994), Stammer (JPO,1997) or Ferrari & Wunsch (2009).

l.40-50 (but also l.309) : when introducing the Mediterranean Sea, remind that the Rossby deformation radius is much smaller in this area, ranging between 20 and 10km. This area is not well described in Chelton et al (1998), but one can also refer to Kurkin et al (https://doi.org/10.2205/2020ES000737, 2020)

Fig.1 : add isolines of EKE to better reflect gradients

l.140 : ‘A focus on the Eastern Alboran gyre’ : not shown

l.147 : For the Ierapetra region : Can you add a description of your region boundary ? It sounds quite small, and therefore may be quite sensitive to shift in boundaries. A more objective box might be drawn taking an EKE-isoline as boundary

l.156 (and after Table.1, l.193) : The ‘area-weighted mean’ was never defined

l.164 : can you add this fitted polynomial to the plot ? If Correlation coeff is so high (0,85), why not using a linear fit instead, as number of satellite is already roughly linear ?

l.171 : The case of the Alboran sea (high EKE but high sensibility) can be a counter-example. Isn’t it rather almost land-locked areas that are more sensitive ? (Alboran, Adriatic, Levantine basin?)

l.188 : about reversal of Ionian Sea circulation, reference about BIOS : Gacic et al (2010)

Table1 : Not clear what are defined as ‘Mean of all trends’ and ‘Trend of the mean’.

l.234 : ‘ability to detect smaller scales features’ This can be asserted by references such as Amores et al (2018) or Stegner et al (2021).

Section3.4 : as described earlier, this subsection is quite interesting but deserves to be investigated further and applied to other regions. A distinction between cyclone and anticyclone should also be made, at least in figure 8. Actually, it sounds surprising that region with cyclonic activity (e.g. South-West Cretean gyre, Rhodes gyre) do not exhibit more sensitivity to increasing number of satellite (in your Fig.5), as they tend be smaller (see observing system experiment in Stegner et al 2021). Do you think EKE trends are rather driven by anticyclone increasing rotating speed over time, as it might be the case for the Alboran gyres ?

Section 4.4 : you can add some discussion also about decennal cycles in the Ionian Sea, where reversal of the regional circulation are expected to have some effects on the EKE of the Midde Mediterranean Jet.

l.275-276 : This behavior could be checked using the META atlas
Citation: https://doi.org/10.5194/egusphere-2025-4651-RC1
- AC1: 'Reply on RC1', Paul Hargous, 19 Dec 2025
  
  We would like to thank the reviewer for devoting their time to revising our manuscript and for their valuable feedback. Please find our comment by comment response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4651-AC1
RC2:
'Comment on egusphere-2025-4651', Pierre-Marie Poulain, 28 Nov 2025

Satellite altimetry products from 1993 to 2023 are assessed to quantify the temporal evolution of surface geostrophic current variability in the Mediterranean Sea. In particular, trends in Eddy Kinetic Energy (EKE) are estimated using three gridded altimetric products. Along-track datasets and an eddy atlas are also employed for comparison in selected regions. Significant positive and negative EKE trends are identified in the most energetic areas of the basin. The influence of the increasing number of satellites on the observed EKE rise is evaluated. Products based on a constant two-satellite configuration appear adequate for estimating long-term trends, but they fail to capture important dynamical structures in some regions. Therefore, the use of constant three-satellite altimetry products is recommended for future investigations of EKE variability in the Mediterranean Sea.
The paper is well structured, and the English is generally fluent and clear. The figures effectively support the written statements. I recommend the publication of this manuscript in Ocean Science, after the authors address the following specific comments (minor revision).
Specific comments:
L43 and L45. Change “waters” to “water”.
L45. Remove “and freshwater”.
L47. Add the following reference before Escudier et al. : Poulain, P. M., Menna, M., & Mauri, E. (2012). Surface geostrophic circulation of the Mediterranean Sea derived from drifter and satellite altimeter data. Journal of Physical Oceanography, 42(6), 973-990. Add this reference in the reference list.
L76. Provide mode details on the Eddy Atlas. Is it based on ADT or SLA structures?
Caption of Fig. 3., L156 and L193. Change “Area-weighted mean” to “Area-averaged”.
Caption of Fig. 6. Change to “Gray-hatched”.
Table 1. Define “Trend of the mean”.
Caption of Fig. 7. Add info about the L3-ref and all-sat-glo curves.
L216. Change “horizontal velocities and their variability are” to “geostrophic velocity variability is”.
L274. Change “However, some go south-east” to “However, some move south-east”.
L306. Change “applications as the two-sat product” to “applications such as the two-sat product”.

Citation: https://doi.org/10.5194/egusphere-2025-4651-RC2
- AC2: 'Reply on RC2', Paul Hargous, 19 Dec 2025
  
  We would like to thank the reviewer for devoting their time to revising our manuscript and for their valuable feedback. Please find our comment by comment response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4651-AC2
RC3:
'Comment on egusphere-2025-4651', Anonymous Referee #3, 09 Dec 2025

The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4651/egusphere-2025-4651-RC3-supplement.pdf

Citation: https://doi.org/10.5194/egusphere-2025-4651-RC3
- AC3: 'Reply on RC3', Paul Hargous, 19 Dec 2025
  
  We would like to thank the reviewer for devoting their time to revising our manuscript and for their valuable feedback. Please find our comment by comment response in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4651-AC3

Peer review completion

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

AR by Paul Hargous on behalf of the Authors (19 Dec 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (19 Dec 2025) by Anne Marie Treguier

RR by Anonymous Referee #1 (26 Dec 2025)

Suggestions for revision or reasons for rejection

The authors answered the referree comments in an adequate way. Of particular interest is the extended analysis in Sect 3.4 about mesoscale eddies distinguishing anticyclones and cyclones. This further highlights the differences between all-sat and two-sat products, but using the framework of coherent mesoscale eddies.

The paragraphs added in the discussion also further explain the observed difference between two-sat and all-sat products. The processes can be either regional pattern changes (Sect4.1) or altimetry processing artefacts (Sect4.2)

Below are some minor specific comments.

l.89 : Is there any filter on eddy lifetime to be considered as a track ?

l.249 "in this section" : remind for the reader that you will use here the META Atlas, as the data source varies between Sect 3.3 and 3.4.

l.253-254 : number of eddies per year is still given adding AE and CE, confusing with Fig.8b

l.247 : Discussing specifically observation of watermasses trapped by mesoscale eddies, one can also refer to Barboni et al (2023) in the Mediterranean Sea, or Laxenaire et al (2019) for Agulhas Rings.

l.303 : "Previous studies have suggested that some
Ierapetra eddies may deviate south-eastward toward the Mersa-Matruh area" you can add that you checked this in a Lagrangian Framework in your supplementary material. It shows that this behavior is quite isolated, hence the EKE trend dipole (Fig.7c) is not a shift of the same mesoscale structure. This seems a fair result to add.

l.373-375 : Nice addition to your conclusion. You can also add (or maybe at the end of discussion) that the same is valid not only for EKE but also to intreprete mesoscale eddies long-term trends. The use of META two-sat product is then advised.

Hide

RR by Anonymous Referee #3 (31 Dec 2025)

ED: Publish subject to minor revisions (review by editor) (05 Jan 2026) by Anne Marie Treguier

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

ED: Publish as is (29 Jan 2026) by Anne Marie Treguier

AR by Paul Hargous on behalf of the Authors (06 Feb 2026)

Journal article(s) based on this preprint

12 Feb 2026

Eddy kinetic energy variability from 30 years of altimetry in the Mediterranean Sea

Paul Hargous, Vincent Combes, Bàrbara Barceló-Llull, and Ananda Pascual

Ocean Sci., 22, 549–564, https://doi.org/10.5194/os-22-549-2026,https://doi.org/10.5194/os-22-549-2026, 2026

Short summary

Paul Hargous, Vincent Combes, Bàrbara Barceló-Llull, and Ananda Pascual

Viewed

Total article views: 680 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
427	212	41	680	24	24

HTML: 427
PDF: 212
XML: 41
Total: 680
BibTeX: 24
EndNote: 24

Views and downloads (calculated since 06 Oct 2025)

Month	HTML	PDF	XML	Total
Oct 2025	137	31	11	179
Nov 2025	97	22	6	125
Dec 2025	90	85	16	191
Jan 2026	72	62	5	139
Feb 2026	31	12	3	46

Cumulative views and downloads (calculated since 06 Oct 2025)

Month	HTML	PDF	XML	Total
Oct 2025	137	31	11	179
Nov 2025	97	22	6	125
Dec 2025	90	85	16	191
Jan 2026	72	62	5	139
Feb 2026	31	12	3	46

Viewed (geographical distribution)

Total article views: 655 (including HTML, PDF, and XML) Thereof 655 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 12 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 (14204 KB)
Metadata XML

Short summary

Over the last three decades, satellites have revealed rising activity of swirling ocean currents in the Mediterranean Sea. We show that the strength of this increase depends strongly on how satellite data are combined. Products that merge many satellites may give the impression of stronger changes simply because coverage improved over time. This work underlines the importance of using stable and consistent datasets to track long-term changes in ocean variability.


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