the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Mar 2025
ENSO-Driven Variability of Deep Ocean Circulation in the Southeast Pacific
Abstract. The Southeast Pacific is crucial for transporting Pacific Deep Water toward the Southern Ocean, primarily through a persistent deep southward flow off the Chilean coast. However, the variability and underlying dynamics of this flow remain poorly understood. Here, we investigate its interannual variability and explore its relationship with El Niño–Southern Oscillation (ENSO). Using the GLORYS12 global ocean reanalysis (1993–2020) and repeated GO-SHIP/WOCE P06 hydrographic sections, we focus on the 2200–4000 m depth range, covering the steep continental slope and the Atacama Trench region. Our analysis confirms a robust southward transport of approximately 2.5 Sv within ~300 km from the coast, predominantly concentrated within 100 km of the continental margin. ENSO exerts a significant influence on this current, with an intensification of the deep southward flow during El Niño events and a weakening during La Niña. This modulation is linked to large-scale oceanic adjustments, including the westward propagation of extratropical Rossby waves, which transfer energy from the surface and thermocline to the deep ocean along steep ray paths consistent with linear theory. Complex Empirical Orthogonal Function (CEOF) analysis and dynamical mode decomposition further reveals that high order baroclinic modes play a key role in shaping deep current variability. Mesoscale processes also seem to contribute to the observed variability. Enhanced deep eddy kinetic energy (EKE) occurs near the continental margin and in the coastal transition zone, which is more prominent during ENSO events. The Analysis of Reynolds stresses and vertical buoyancy fluxes (VBF) indicates that energy is transferred between mean flow and turbulence, with VBF being a dominant contributor to deep variability of EKE tendency. These findings emphasize the complex interplay of large-scale climatic forcing, topography, and mesoscale turbulence in shaping deep ocean circulation. Future work should integrate sustained deep-ocean observations and high-resolution modelling to refine our understanding of the DESPC and its role in the Pacific Meridional Overturning Circulation.
- Preprint
(3059 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed (peer review stopped)
- RC1: 'Comment on egusphere-2025-1311', Anonymous Referee #1, 29 Apr 2025
-
RC2: 'Comment on egusphere-2025-1311', Anonymous Referee #2, 04 Jun 2025
Review egusphere-2025-1311
ENSO-driven variability of deep ocean circulation in the southeast Pacific by Manuel Torres-Godoy, Oscar Pizarro, Boris Dewitte and Vera Oerder
This manuscript investigates the variability of the deep southward current and transport along the eastern Pacific margin adjacent to the Chilean coast. The study concentrates on assessing the velocity and transport variability within 300-100 kms of the coast and explores the relationship with El Niño-Southern Oscillation (ENSO). The analysis is undertaken using the GLORYS12 global ocean reanalysis (1993-2020). There is extensive use Hamming filter of various temporal width (19 months, 101 days and 25 days) on the seasonal anomaly velocity time series. The intraseasonal time series is calculated as the residual of the low frequency (101 day) time-series and the high frequency (<25 day) time-series. The intraseasonal time-series is then used to calculate the various terms of the Eddy Kinetic Energy budget.
While that the topic may be of interest to the ocean science community and provide new insights into the magnitude and drivers of deep ocean variability, the current manuscript has significant issues regarding developing and justifying the science topic and presenting these results in a clear and concise, and well-structured manner. In addition, the methods and application were not provided with appropriate detail to enable the reader to fully understand and or determine that they were appropriate. For these reasons I find that the manuscript, as presented, is not acceptable for publication in Ocean Science. I would encourage the authors to re-consider the methods used, analysis and presentation of some on the results in a newly constructed manuscript.
Major Comments
The Introduction points to the major role played by southward flowing Pacific Deep Water in the global overturning circulation. While this is accepted, the importance of the narrow deep current adjacent to the Chilean coast as a component of the Pacific wide or even eastern South Pacific southward flow of Pacific Deep Water is unclear. How does the persistence and variability of the deep flow at the eastern boundary influence the southward flow of the Pacific Deep Water and therefore the structure of the Pacific Deep overturing circulation?
The primary aim of the manuscript is to provide improved understanding of the “variability of the deep southward flow in the Chile Basin and its relationship with El Niño cycles” and the assessment of the dynamics that mediate this relationship. Given this, it is unclear why such a large amount of the introduction is dedicated to describing the large-scale circulation of Pacific Deep Water.
The Material and Methods section is not well presented; important analysis methods lack detail, and it is unclear why some sections are included. For example, the description of the study area is confusing. Upon reading the manuscript it is unclear why the larger region of the eastern Pacific Ocean is discussed, including the use of WOCE/GO-SHIP P06 section and figure 2. It is unclear why a comparison of GLORYS to P06 is provided.
In section 2.2 it is stated that “our analysis focuses on depths from roughly 2000 m to 4000 m” and the 6-depth levels of GLORYS are articulated, yet many of the manuscript figures and discussion/interpretation thereof extend to the surface or about 1000m.
The methods and derivation of the terms in the Eddy Kinetic Energy budget is not provided at an appropriate level of detail in section 2.4. Various Hamming filter widths (19 months, 25 days and 101days) are applied to the velocity anomaly time-series, of which the residual of the 101 day and 25 days anomaly time-series is used as the intraseasonal time-series. Would it be more appropriate to have used a band-pass filter of appropriate width? A clear, accurate description of the methods applied to the velocity anomaly time series is vital, as the intraseasonal anomaly time series is used to calculate the terms of the Eddy Kinetic Energy budget.
Lines 224-228 – explaining relationship of thermocline perturbations, ENSO and Rossby waves is more appropriate in the results sections when these connections/concepts are required.
Finally, it is unclear why sections 2.3 and 2.4 are provided. The hydrographic data are not required for this study and the CEOF information is not central to the science results presented in the manuscript. The description of CEOF at lines 373-376 is all that is required.
The structure and order of presentation of the results section confuses the reader. It is unclear what is the main message of sections 3.1 and 3.2 and how this is connected to understanding the variability of the narrow deep boundary current and connection to ENSO. What is the value of figure 2 and associated discussion? How does the low-frequency anomalous velocity ENSO composite analysis link with or lead to the intraseasonal analysis of the terms of the Eddy Kinetic Energy budget? In section 3.2 is the velocity time-series SIA is used? Or do you composite the SA time series? What new information is shown in Figure 5 that is not already shown in Figure 4?
It is suggested that author consider remove section 3.1 and retain only that which is essential for the manuscript. Also, it is suggested that components of section 3.2 be removed and other components linking ENSO to deep ocean variability be moved to after section 3.3.
The Discussion section 4.1, while of interest is not the main topic of this manuscript and could be removed. In general, the Discussion paragraphs are a summary of the figures and results sections, and there is limited comparison with previous studies and identification of the new knowledge/information provided by this study.
Citation: https://doi.org/10.5194/egusphere-2025-1311-RC2 -
EC1: 'Comment on egusphere-2025-1311', Katsuro Katsumata, 05 Jun 2025
We have two reviewers' reports. Since the opinions are mixed, I have asked a third reviewer for comments. Please wait until the comments are posted. We thank you for your patience.
Citation: https://doi.org/10.5194/egusphere-2025-1311-EC1 -
RC3: 'Comment on egusphere-2025-1311', Anonymous Referee #3, 18 Jun 2025
Review of ENSO-driven variability of deep ocean circulation in the Southeast Pacific by Torres-Godoy et al.
General comments
The authors describe the deep currents off Chile mainly using reanalysis data. However, the writing is very poor. Their explanation is not straightforward and not sufficient. Hence, I am afraid that I do not understand their conclusions fully.
This article is composed of three topics. The first topic (3.1) is the southward current as global thermohaline circulation. The authors compare the velocities of the reanalysis data with those of hydrographic observations. Although I do not know GLORYS12, it targets the upper ocean and it would not be suitable for a study of the thermohaline circulation. I have some questions. Does the modeled deep ocean reach a quasi-steady state? Does the modeled hydrography agree with observation (this depends on what data are assimilated)? Does the deep water off Chile truly originate from the North Pacific? If the authors still want to discuss this topic, the authors should examine both of currents and water mass in the whole Southeast Pacific (not only a narrow area off Chile).
The second topic (3.2) is the variability coherent to the ENSO. This is interesting and an important finding for the deep circulation. Unfortunately, the authors examine it only in a narrow area off Chile. When the southward current is strong, what happens in the upstream and downstream regions? The reanalysis data can show it. It would be interesting to make a horizontal map of correlation in the whole Pacific in such a way as Figure 5. To confirm the variability, the authors should also examine the difference in four hydrographic observations at P06 (I do not know their timings in the ENSO cycle). I do not understand the Rossby wave and CEOF results. The authors may suggest that Rossby waves cause deep variability, but Figure 6 shows no apparent relation to the ENSO for me. They propagate westward so slowly that they do not cross the study area in one ENSO cycle. For the CEOF analysis, the authors state that no modes correlate the ENSO. This means the ENSO-related variability is less significant than the other ones.
The third topic (3.3) is the EKE. I do not understand what the authors conclude. The high EKE is outstanding near 35S, 78W. However, the authors only examine the coastal region (I do not know where the CTZ is). They want to explain the time change of EKE by Reynolds stress, but the VRS and HRS are much smaller than VBF. It would be suggested that baroclinic instabilities occur. More importantly, they ignore the advection of EKE. Because the equation in L180 has a material derivative, the advection is necessary on the analysis on a fixed point. Ordinarily, all the terms including time change are displayed to show their exact balance. In addition, the relation of EKE to the Rossby waves in 3.2 is not clear. Is the EKE elevated by the wave through Reynolds stress?
Accordingly, I do not recommend the publication of this article. However, the topic of 3.2 is very worth exploring. I hope that the authors will make their findings more scientifically confirmable and submit a new manuscript.
Specific comments
Please consider these comments in a future manuscript.
L121: Their levels are down only to 5700m. How well is the trench resolved in the model?
L139, Figure 2, Table 2: Is the determination of no motion level consistent with ENSO-related variability? What reference is used in western shallower regions? Bottom-referenced?
L154: There are many filtered time series. In Section 3.1 and 3.2, I do not realize which time series is used.
L216: In Section 3, the ENSO periods are used. Which criterion is used, moderated or strong (except for Figure 3)?
L340: First vertical mode is ambiguous. Write first baroclinic mode.
L360: I do not identify the upward propagation in Figure 7. Draw an indicator as well as in Figure 6.
L365: Write the reference which shows that upward phase means downward energy.
Figure 5: Correlation and regression coefficient are redundant because both are derived from covariance.
Figure 9: Phase propagation can be displayed concisely by iso-phase lines. This figure is redundant (at lease, panels from 180 to 315 degrees).
Figure 10: GLORYS12 extends only to 5700m. Why is 6000m isobath drawn?
Table 1: What does the “error” mean in the caption, standard deviation or standard error? The difference between values in El Nino and La Nina is much smaller than the error. Is it statistically significant?
Citation: https://doi.org/10.5194/egusphere-2025-1311-RC3 -
EC2: 'Comment on egusphere-2025-1311', Katsuro Katsumata, 14 Jul 2025
Although the comments from Referee #1 indicate a possibility of revision, the other two Referees have found serious issues in the method and presentation. Given these comments and based on my reading, I must discourage submission of a revised manuscript. The three reports include suggestions for improvements. The author may choose to re-submit the mansucript after attending to these comments, but it will be treated as a new submission.
Citation: https://doi.org/10.5194/egusphere-2025-1311-EC2 -
AC1: 'Reply on EC2', Oscar Pizarro, 27 Jul 2025
Thank you for your message and for the time and effort you and the reviewers have dedicated to evaluating our manuscript. We sincerely appreciate the feedback provided by all three referees. We acknowledge the concerns raised regarding the methodology and presentation, and we understand the editorial decision. Taking the reviewers’ and your suggestions into careful consideration, we plan to substantially revise our work and prepare a new manuscript that addresses the main issues highlighted in the reviews.
We thank you once again for your time and guidance.
Citation: https://doi.org/10.5194/egusphere-2025-1311-AC1
-
AC1: 'Reply on EC2', Oscar Pizarro, 27 Jul 2025
Status: closed (peer review stopped)
- RC1: 'Comment on egusphere-2025-1311', Anonymous Referee #1, 29 Apr 2025
-
RC2: 'Comment on egusphere-2025-1311', Anonymous Referee #2, 04 Jun 2025
Review egusphere-2025-1311
ENSO-driven variability of deep ocean circulation in the southeast Pacific by Manuel Torres-Godoy, Oscar Pizarro, Boris Dewitte and Vera Oerder
This manuscript investigates the variability of the deep southward current and transport along the eastern Pacific margin adjacent to the Chilean coast. The study concentrates on assessing the velocity and transport variability within 300-100 kms of the coast and explores the relationship with El Niño-Southern Oscillation (ENSO). The analysis is undertaken using the GLORYS12 global ocean reanalysis (1993-2020). There is extensive use Hamming filter of various temporal width (19 months, 101 days and 25 days) on the seasonal anomaly velocity time series. The intraseasonal time series is calculated as the residual of the low frequency (101 day) time-series and the high frequency (<25 day) time-series. The intraseasonal time-series is then used to calculate the various terms of the Eddy Kinetic Energy budget.
While that the topic may be of interest to the ocean science community and provide new insights into the magnitude and drivers of deep ocean variability, the current manuscript has significant issues regarding developing and justifying the science topic and presenting these results in a clear and concise, and well-structured manner. In addition, the methods and application were not provided with appropriate detail to enable the reader to fully understand and or determine that they were appropriate. For these reasons I find that the manuscript, as presented, is not acceptable for publication in Ocean Science. I would encourage the authors to re-consider the methods used, analysis and presentation of some on the results in a newly constructed manuscript.
Major Comments
The Introduction points to the major role played by southward flowing Pacific Deep Water in the global overturning circulation. While this is accepted, the importance of the narrow deep current adjacent to the Chilean coast as a component of the Pacific wide or even eastern South Pacific southward flow of Pacific Deep Water is unclear. How does the persistence and variability of the deep flow at the eastern boundary influence the southward flow of the Pacific Deep Water and therefore the structure of the Pacific Deep overturing circulation?
The primary aim of the manuscript is to provide improved understanding of the “variability of the deep southward flow in the Chile Basin and its relationship with El Niño cycles” and the assessment of the dynamics that mediate this relationship. Given this, it is unclear why such a large amount of the introduction is dedicated to describing the large-scale circulation of Pacific Deep Water.
The Material and Methods section is not well presented; important analysis methods lack detail, and it is unclear why some sections are included. For example, the description of the study area is confusing. Upon reading the manuscript it is unclear why the larger region of the eastern Pacific Ocean is discussed, including the use of WOCE/GO-SHIP P06 section and figure 2. It is unclear why a comparison of GLORYS to P06 is provided.
In section 2.2 it is stated that “our analysis focuses on depths from roughly 2000 m to 4000 m” and the 6-depth levels of GLORYS are articulated, yet many of the manuscript figures and discussion/interpretation thereof extend to the surface or about 1000m.
The methods and derivation of the terms in the Eddy Kinetic Energy budget is not provided at an appropriate level of detail in section 2.4. Various Hamming filter widths (19 months, 25 days and 101days) are applied to the velocity anomaly time-series, of which the residual of the 101 day and 25 days anomaly time-series is used as the intraseasonal time-series. Would it be more appropriate to have used a band-pass filter of appropriate width? A clear, accurate description of the methods applied to the velocity anomaly time series is vital, as the intraseasonal anomaly time series is used to calculate the terms of the Eddy Kinetic Energy budget.
Lines 224-228 – explaining relationship of thermocline perturbations, ENSO and Rossby waves is more appropriate in the results sections when these connections/concepts are required.
Finally, it is unclear why sections 2.3 and 2.4 are provided. The hydrographic data are not required for this study and the CEOF information is not central to the science results presented in the manuscript. The description of CEOF at lines 373-376 is all that is required.
The structure and order of presentation of the results section confuses the reader. It is unclear what is the main message of sections 3.1 and 3.2 and how this is connected to understanding the variability of the narrow deep boundary current and connection to ENSO. What is the value of figure 2 and associated discussion? How does the low-frequency anomalous velocity ENSO composite analysis link with or lead to the intraseasonal analysis of the terms of the Eddy Kinetic Energy budget? In section 3.2 is the velocity time-series SIA is used? Or do you composite the SA time series? What new information is shown in Figure 5 that is not already shown in Figure 4?
It is suggested that author consider remove section 3.1 and retain only that which is essential for the manuscript. Also, it is suggested that components of section 3.2 be removed and other components linking ENSO to deep ocean variability be moved to after section 3.3.
The Discussion section 4.1, while of interest is not the main topic of this manuscript and could be removed. In general, the Discussion paragraphs are a summary of the figures and results sections, and there is limited comparison with previous studies and identification of the new knowledge/information provided by this study.
Citation: https://doi.org/10.5194/egusphere-2025-1311-RC2 -
EC1: 'Comment on egusphere-2025-1311', Katsuro Katsumata, 05 Jun 2025
We have two reviewers' reports. Since the opinions are mixed, I have asked a third reviewer for comments. Please wait until the comments are posted. We thank you for your patience.
Citation: https://doi.org/10.5194/egusphere-2025-1311-EC1 -
RC3: 'Comment on egusphere-2025-1311', Anonymous Referee #3, 18 Jun 2025
Review of ENSO-driven variability of deep ocean circulation in the Southeast Pacific by Torres-Godoy et al.
General comments
The authors describe the deep currents off Chile mainly using reanalysis data. However, the writing is very poor. Their explanation is not straightforward and not sufficient. Hence, I am afraid that I do not understand their conclusions fully.
This article is composed of three topics. The first topic (3.1) is the southward current as global thermohaline circulation. The authors compare the velocities of the reanalysis data with those of hydrographic observations. Although I do not know GLORYS12, it targets the upper ocean and it would not be suitable for a study of the thermohaline circulation. I have some questions. Does the modeled deep ocean reach a quasi-steady state? Does the modeled hydrography agree with observation (this depends on what data are assimilated)? Does the deep water off Chile truly originate from the North Pacific? If the authors still want to discuss this topic, the authors should examine both of currents and water mass in the whole Southeast Pacific (not only a narrow area off Chile).
The second topic (3.2) is the variability coherent to the ENSO. This is interesting and an important finding for the deep circulation. Unfortunately, the authors examine it only in a narrow area off Chile. When the southward current is strong, what happens in the upstream and downstream regions? The reanalysis data can show it. It would be interesting to make a horizontal map of correlation in the whole Pacific in such a way as Figure 5. To confirm the variability, the authors should also examine the difference in four hydrographic observations at P06 (I do not know their timings in the ENSO cycle). I do not understand the Rossby wave and CEOF results. The authors may suggest that Rossby waves cause deep variability, but Figure 6 shows no apparent relation to the ENSO for me. They propagate westward so slowly that they do not cross the study area in one ENSO cycle. For the CEOF analysis, the authors state that no modes correlate the ENSO. This means the ENSO-related variability is less significant than the other ones.
The third topic (3.3) is the EKE. I do not understand what the authors conclude. The high EKE is outstanding near 35S, 78W. However, the authors only examine the coastal region (I do not know where the CTZ is). They want to explain the time change of EKE by Reynolds stress, but the VRS and HRS are much smaller than VBF. It would be suggested that baroclinic instabilities occur. More importantly, they ignore the advection of EKE. Because the equation in L180 has a material derivative, the advection is necessary on the analysis on a fixed point. Ordinarily, all the terms including time change are displayed to show their exact balance. In addition, the relation of EKE to the Rossby waves in 3.2 is not clear. Is the EKE elevated by the wave through Reynolds stress?
Accordingly, I do not recommend the publication of this article. However, the topic of 3.2 is very worth exploring. I hope that the authors will make their findings more scientifically confirmable and submit a new manuscript.
Specific comments
Please consider these comments in a future manuscript.
L121: Their levels are down only to 5700m. How well is the trench resolved in the model?
L139, Figure 2, Table 2: Is the determination of no motion level consistent with ENSO-related variability? What reference is used in western shallower regions? Bottom-referenced?
L154: There are many filtered time series. In Section 3.1 and 3.2, I do not realize which time series is used.
L216: In Section 3, the ENSO periods are used. Which criterion is used, moderated or strong (except for Figure 3)?
L340: First vertical mode is ambiguous. Write first baroclinic mode.
L360: I do not identify the upward propagation in Figure 7. Draw an indicator as well as in Figure 6.
L365: Write the reference which shows that upward phase means downward energy.
Figure 5: Correlation and regression coefficient are redundant because both are derived from covariance.
Figure 9: Phase propagation can be displayed concisely by iso-phase lines. This figure is redundant (at lease, panels from 180 to 315 degrees).
Figure 10: GLORYS12 extends only to 5700m. Why is 6000m isobath drawn?
Table 1: What does the “error” mean in the caption, standard deviation or standard error? The difference between values in El Nino and La Nina is much smaller than the error. Is it statistically significant?
Citation: https://doi.org/10.5194/egusphere-2025-1311-RC3 -
EC2: 'Comment on egusphere-2025-1311', Katsuro Katsumata, 14 Jul 2025
Although the comments from Referee #1 indicate a possibility of revision, the other two Referees have found serious issues in the method and presentation. Given these comments and based on my reading, I must discourage submission of a revised manuscript. The three reports include suggestions for improvements. The author may choose to re-submit the mansucript after attending to these comments, but it will be treated as a new submission.
Citation: https://doi.org/10.5194/egusphere-2025-1311-EC2 -
AC1: 'Reply on EC2', Oscar Pizarro, 27 Jul 2025
Thank you for your message and for the time and effort you and the reviewers have dedicated to evaluating our manuscript. We sincerely appreciate the feedback provided by all three referees. We acknowledge the concerns raised regarding the methodology and presentation, and we understand the editorial decision. Taking the reviewers’ and your suggestions into careful consideration, we plan to substantially revise our work and prepare a new manuscript that addresses the main issues highlighted in the reviews.
We thank you once again for your time and guidance.
Citation: https://doi.org/10.5194/egusphere-2025-1311-AC1
-
AC1: 'Reply on EC2', Oscar Pizarro, 27 Jul 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 383 | 79 | 24 | 486 | 15 | 34 |
- HTML: 383
- PDF: 79
- XML: 24
- Total: 486
- BibTeX: 15
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Please see my comments in the attached PDF file.