Inclination Trend of the Agulhas Return Current Path in Three Decades

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Review of “Inclination trend of the Agulhas Return Current path in three decades”.

This paper reveals that the ARC axis exhibits a significant slanting trend with its western part migrating northward and the eastern part migrating southward in 1993-2020. The Ekman pumping induced by wind stress anomalies leads to the tilt of the ARC. In my mind, the paper is generally self-consistent and well-written. But there are some issues are not clear to me. I think this paper needs minor revision before publish in Remote sensing. The comments are as follows:

Line 32-36：Add more reference. 

Line 58: The variability of the Agulhas retroflection can also have an impact on the local climate (Zhu et al. 2021). 

Zhu, Y., Li, Y., Zhang, Z., Qiu, B., & Wang, F. (2021). The observed Agulhas Retroflection behaviors during 1993–2018. Journal of Geophysical Research: Oceans, 126, e2021JC017995.

2.3 Definition of the ARC Jet Axis: you choose different SSH contour to define the ARC in the west and east of 50E, and the reasons for such a choice are not clearly described here. You can give an example as well as a figure to help make it clearer. By the way, why is 50E? Additionally, the authors have want to cite that Fadida, et al. found the variability of the ARC is different in the east and west parts of the ARC. What is the difference between your two ARC definition methods?

Line 143: It is better to describe how the normalization function calculated. 

Figure 3: Please better explain the tilt of the ARC. What does the ARC look like when the AMI index in the high and low phases, respectively?  

Line 199: I do not see how EKE is calculated in the paper. 

Line 226: To quantify this, the authors may provide the correlation coefficient between Figure 5 and Figure 6. 

Comments on the Quality of English Language

The writing can be improved.

Author Response

This paper reveals that the ARC axis exhibits a significant slanting trend with its western part migrating northward and the eastern part migrating southward in 1993-2020. The Ekman pumping induced by wind stress anomalies leads to the tilt of the ARC. In my mind, the paper is generally self-consistent and well-written. But there are some issues are not clear to me. I think this paper needs minor revision before publish in Remote sensing. The comments are as follows:

Line 32-36：Add more reference. 

R： Revised.

Line 58: The variability of the Agulhas retroflection can also have an impact on the local climate (Zhu et al. 2021). 

Zhu, Y., Li, Y., Zhang, Z., Qiu, B., & Wang, F. (2021). The observed Agulhas Retroflection behaviors during 1993–2018. Journal of Geophysical Research: Oceans, 126, e2021JC017995.

 

R: Revised. Thanks for the information!

 

 

2.3 Definition of the ARC Jet Axis: you choose different SSH contour to define the ARC in the west and east of 50E, and the reasons for such a choice are not clearly described here. You can give an example as well as a figure to help make it clearer.

By the way, why is 50E? Additionally, the authors have want to cite that Fadida, et al. found the variability of the ARC is different in the east and west parts of the ARC. What is the difference between your two ARC definition methods?

 

R: Thank you for your valuable suggestion!  

Firstly, we choose different SSH contour to define the ARC axis because single ADT contour cannot perfectly fit the ADT gradient maximum throughout the whole jet path. The upstream ARC is mainly constrained by topography and relatively steady. But in the downstream the ARC becomes weaker and more prone to vibration, leading to the greater deviation between the ADT contour and maximum geostrophic velocity. We added Figure 1 in the revised manuscript to illustrate our method. For example, the climatological 0.7 m ADT contour west of 50°E is very close to the geostrophic velocity maximum while east of 50°E it is farther north (Figure. 1a). It seems that the upstream ARC (approximately 20°E~50°E) is mainly constrained to bathymetry, whereas the downstream (approximately 50°E~70°E) weakens and varies greatly. In general, the maximum ADT gradients mostly locate within the range of 0.6m ~ 0.9m ADT contours west of 50°E and 0.5m ~ 0.7m ADT contours east of 50°E.

Secondly, the dividing line 50°E is not selected based on a strict standard but just by visual. The line can be shifted slightly eastward or westward near the 50°E without compromising the integrity of the ARC nor qualitatively changing the results. More detailed explanation provided in the section 2.3 in the revised version.

In comparison to Fadida et al., we underline the different features of ARC path in the west and east sections. In fact, the downstream ARC is much weaker and expose to a complex circulation background. It is closer to the subtropical front and more susceptible to its influence. Therefore, we applied different criteria for upstream and downstream sections in identifying the ARC axis position. Our method is superior to Fadida et al in identifying the ARC jet axis from the perspective of dynamics. In their study, the maximum SSH gradient along each longitudinal grid is searched within a fixed area, i.e., 36.125-46.125°S and 20.125-70.325°E. Since the downstream jet is less bathymetric restrained and gets closer to the ACC fronts, multiple maximums and their interference occur frequently in this section. Our method, by restricting the search area along the specific ADT contours, obviously improves the identification of ARC jet axis. The resulting ARC axis slanting tendency in this study is more owing to the substantial southward shift of the downstream area and less to the upstream. In particular, the tendency of the headstream section west of 35°E are almost neutral and insignificant, which shows a great discrepancy with Fadida et al.

 

 

Line 143: It is better to describe how the normalization function calculated. 

R: We add the normalization function computation in Section 3.1: “zscore is normalization function, calculated by the formula zscore(x)=(x-μ)/σ, where μ is the x mean and σ is the x standard deviation.”

 

Figure 3: Please better explain the tilt of the ARC. What does the ARC look like when the AMI index in the high and low phases, respectively?  

R: According to your suggestion, we add the figure (Figure 4) and related descriptions in Section 3.1.

 

Line 199: I do not see how EKE is calculated in the paper. 

R: The formula for computing EKE is added in the text.

 

Line 226: To quantify this, the authors may provide the correlation coefficient between Figure 5 and Figure 6. 

R: Thanks for your suggestion! We compute the correlation coefficient and provided the values in the section 3.3.

Reviewer 2 Report

Comments and Suggestions for Authors

The Agulhas Return Current (ARC), a vital component of the Agulhas system, plays a crucial role in the exchange of water and material between the southern regions of the Indian and Atlantic Ocean basins. In this study, the authors utilize Satellite-based Absolute Dynamic Topography and ERA5 reanalysis wind dataset to discuss the changing trends in the ARC's position. The study reveals that under the influence of wind stress curl, the ARC axis exhibits a significant slanting trend, with its western part migrating northward and the eastern part migrating southward. Furthermore, the paper explores the impact of the ARC's main axis movement on eddy kinetic energy. The chosen topic is interesting, but the paper mainly consists of conceptual discussions, lacking essential quantification, which affects the value of the results. Additionally, the paper is relatively concise, reflected not only in the limited use of seven figures but also in its overall length, which is less than half the typical length of common publications in Remote Sensing (approximately 18 pages). This brevity gives the impression of insufficient content. Below are some suggestions for revision, which the authors may consider.

General Comments：

1. In this paper, the enhancement and weakening of wind stress curl are used to indicate that wind stress curl can induce vertical motion in the subsurface layer through Ekman suction. However, further quantification and explanation are needed to understand the relationship between vertical motion and the movement of the ARC's position.

2. It is suggested to appropriately increase the comparison between the research results of this paper and those of Fadida, et al. Although the authors have emphasized the potential mechanisms behind the changes in the ARC's main axis in this study, a more comprehensive comparison might be necessary, given the somewhat limited length of this paper.

3. In the specific determination of the ARC's main axis, due to the precision of satellite observation data and the influence of complex oceanic conditions, is it possible to encounter significant deviations in the latitude positions of the calculated ARC between two adjacent longitudes? It is recommended that the authors provide a more detailed explanation of the specific calculation method for the maximum gradient of Absolute Dynamic Topography (ADT). This would aid readers in replicating this work.

4. Line 149: “It is notable that the ARC path as a whole exhibits a strong seasonal cycle of north-south migration with its shape intact (figure not shown)”. Considering the relatively short of the article, it is advisable for the authors to include the corresponding figure and add a section discussing the seasonal characteristics of the ARC's position, along with its long-term trends.

Minor Comments:

1. Line 39: “… transport as well as filament advection [1,2].” The study by Wei and Wang, 2023, provides a detailed discussion on Agulhas leakage, which could be a valuable reference for the authors.

Reference:

Wei L, Wang C. Characteristics of ocean mesoscale eddies in the Agulhas and Tasman Leakage regions from two eddy datasets. Deep Sea Research Part II: Topical Studies in Oceanography. (2023) 208: 105264. doi: 10.1016/j.dsr2.2023.105264

2. In some parts of the article, the term “downstream” refers to east of 50°E, while in other instances, it refers to east of 48°E. It is recommended to standardize the terminology for consistency throughout the article.

3. Line 115: "While east of 50°E, the search area is within ± 0.7° of the latitude." What is the rationale behind choosing 0.7° in this context?

4. Lines 162-165: “Since the spatial distribution of climatological ADT is high in the north and low in the south within the ARC region, the tripolar positive-negative-positive pattern of ADT anomalies in the western section indicates that the geostrophic velocity enhances on the north side of the ARC jet axis and reduces on the south side.” Since the anomalies in ADT are positive on both the north and south sides of the ARC, why does the geostrophic velocity strengthen on the north side and weaken on the south side of the ARC jet axis? Could the authors provide a more detailed explanation for this phenomenon?

5. Line 198, (unit in N/m3) -->  (unit in N/m³), and also in Line 219

6. Lines 231 ~ 233 “The headstream EKE trends, on the other side, may be more susceptible to other factors such as topographical meandering rather than the ARC movement” This statement appears to be a speculation. It would be advisable for the authors to provide further explanation or evidence to support this claim.

 7. How is Eddy Kinetic Energy defined in the study? How are the anomalous velocity values calculated?

Author Response

General Comments：

1. In this paper, the enhancement and weakening of wind stress curl are used to indicate that wind stress curl can induce vertical motion in the subsurface layer through Ekman suction. However, further quantification and explanation are needed to understand the relationship between vertical motion and the movement of the ARC's position.

R: Thank you for your advice. We provid more detailed description of this mechanism in section 3.1 L269-284. The wind stress curl anomaly is moderately negative in the center of the ARC jet axis and substantially positive on the north and south sides in western section. This wind stress curl pattern may induce anomalous vertical motion via Ekman pumping effect, i.e., upwelling in the center of jet axis and downwelling at both flanks in the western section. The meridional tripolar vertical motion anomalies probably trigger the dynamical adjustment of both the sea surface height and the subsurface isopycnals, accounting for the positive-negative- positive ADT anomalies in the western section.

 

2. It is suggested to appropriately increase the comparison between the research results of this paper and those of Fadida, et al. Although the authors have emphasized the potential mechanisms behind the changes in the ARC's main axis in this study, a more comprehensive comparison might be necessary, given the somewhat limited length of this paper.

R: Thank you for your insightful comments! Our study is in agreement with Fadida et al. to confirm the slanting trends of the ARC axis in general.

However, our method to identify the ARC axis is different to their study. They searched the maximum SSH gradient within a fixed area (36.125-46.125°S and 20.125-70.325°E) while we choose to confine the searching area within a range of ADT contour. Furthermore, we adopt different criteria for the search ranges of upstream and downstream regions in the light of their different kinematic characteristics and circulation background.

By adopting this new method, we reveal that the ARC slanting trend is more dependent on the southward shift of the downstream axis and less on the topographic-steering upstream. It is worth noting that in the headstream, the ARC axis trends manifest as the deformation of its large meanders instead of integral northward migration. This is totally different to the result of Fadida et al.

In addition, the mechanisms leading to the slanting trend of ARC axis and the deformation trend in the head stream and their connections with the local eddy activity are discussed in detail in this study. In contrast, Fadida et al. lacks the detailed depiction of physical mechanism, though they mentioned the possible connection between the slanting of ARC axis, wind forcing and EKE.

We follow your suggestion to added more analysis (Figure 10-15 and the related text content) and present a more comprehensive comparison with Fadida et al. in section 2.3, 3.1 and section 4.

 

3. In the specific determination of the ARC's main axis, due to the precision of satellite observation data and the influence of complex oceanic conditions, is it possible to encounter significant deviations in the latitude positions of the calculated ARC between two adjacent longitudes? It is recommended that the authors provide a more detailed explanation of the specific calculation method for the maximum gradient of Absolute Dynamic Topography (ADT). This would aid readers in replicating this work.

R: Since the definition is a combination of two methods and the search area is limited, no significant deviations have been found between two adjacent longitudes. We provide a detailed explanation of the specific calculation method of maximum gradient of ADT in section 2.3 in the revised manuscript.

4. Line 149: “It is notable that the ARC path as a whole exhibits a strong seasonal cycle of north-south migration with its shape intact (figure not shown)”. Considering the relatively short of the article, it is advisable for the authors to include the corresponding figure and add a section discussing the seasonal characteristics of the ARC's position, along with its long-term trends.

R: We add figure 5 and the related description in revised version to present the seasonal cycle of ARC path.

 

Minor Comments:

1. Line 39: “… transport as well as filament advection [1,2].” The study by Wei and Wang, 2023, provides a detailed discussion on Agulhas leakage, which could be a valuable reference for the authors.

Reference:

Wei L, Wang C. Characteristics of ocean mesoscale eddies in the Agulhas and Tasman Leakage regions from two eddy datasets. Deep Sea Research Part II: Topical Studies in Oceanography. (2023) 208: 105264. doi: 10.1016/j.dsr2.2023.105264

R: The reference is added now in the section 1. Thank you for the information!

 

2. In some parts of the article, the term “downstream” refers to east of 50°E, while in other instances, it refers to east of 48°E. It is recommended to standardize the terminology for consistency throughout the article.

R: In identifying the ARC axis, we choose 50°E as the border of upstream and downstream. This longitude is selected by visual, not on a strict standard. The line can be shifted slightly eastward or westward near the 50°E without compromising the integrity of the ARC nor qualitatively changing the results. However, by computing the axis trends, the significant northward and southward shift of ARC path occurs on the east and west sides of 48°E. Therefore, we finally adopt the 48°E as the dividing line. We add more words to explain this process in section 2.3, in order to avoid possible misunderstanding.

 

3. Line 115: "While east of 50°E, the search area is within ± 0.7° of the latitude." What is the rationale behind choosing 0.7° in this context?

R: Thanks for your comments. The ± 0.7°range of searching area is chosen because the downstream ARC is placed in a complex circulation background. On one hand it displays a large variability after detaching from the strict terrain constraints, thus the search area cannot be too narrow. On the other hands it is exposed to the risk of ACC frontal and jets interference if the searching scope is excessive large. We need a compromise between them. After trying multiple times, we choose 0.7°as the balanced level, and the resulting downstream axis in the eastern section can better interface with the western section in spite of different criteria are adopted.

 

4. Lines 162-165: “Since the spatial distribution of climatological ADT is high in the north and low in the south within the ARC region, the tripolar positive-negative-positive pattern of ADT anomalies in the western section indicates that the geostrophic velocity enhances on the north side of the ARC jet axis and reduces on the south side.” Since the anomalies in ADT are positive on both the north and south sides of the ARC, why does the geostrophic velocity strengthen on the north side and weaken on the south side of the ARC jet axis? Could the authors provide a more detailed explanation for this phenomenon?

R: Since the climatological ADT gradient points north (ADT values increasing from south to north), the negative ADT anomaly at ARC axis and positive ADT anomaly at its north means the increase of local meridional ADT gradient, i.e., ADT contours become closer. This corresponds to the strengthening of the geostrophic velocity on the north side. Similarly, the negative ADT anomaly at ARC axis and positive ADT anomaly at its south means the decrease of ADT gradient, thus the weakening on the south side.

 

5. Line 198, (unit in N/m3) -->  (unit in N/m³), and also in Line 219

R: Revised. Thank you very much!

 

6. Lines 231 ~ 233 “The headstream EKE trends, on the other side, may be more susceptible to other factors such as topographical meandering rather than the ARC movement” This statement appears to be a speculation. It would be advisable for the authors to provide further explanation or evidence to support this claim.

R: Thank you for your invaluable suggestion! We add more analyses and figures in the revised manuscript to support this statement. See section 3.3.

 

7. How is Eddy Kinetic Energy defined in the study? How are the anomalous velocity values calculated?

R: We add the EKE definition in section 2.4.

Reviewer 3 Report

Comments and Suggestions for Authors

In their paper entitled “Inclination trend of the Agulhas Return Current path in three 3 decades”, Lin et al. use the AVISO merged altimetry product as well wind data from the ERA5 reanalysis to investigate long term changes in the Agulhas Return Current and its potential causes. Their results show that the path of the Agulhas Return Current has changed with a tilting of the current’s axis due to a northward movement of the current in its eastern section and a southward movement of the current in its western section. Lin et al. attribute the tilting of the Agulhas Return Current to large scale changes in the wind stress curl.

While the paper is well written, structured and the analysis supports the results, it is not clear to me how the results differ from those published by Fatida et al. (2021). The altimetry data used by Lin et al. is  the same as that used by Fatida et al. (2021). In their paper, Fatida et al. (2021) also report on the same slanting trend of the Agulhas Return Current and point to the wind stress curl as a driving mechanism. The analysis of Fatida et al. (2021) in fact goes further than the results presented here by Lin et al., as Fatida and al. (2021) looks at the seasonal trends and variability in the trends over the period of analysis. The paper of Fatida et al. (2021) also looks at the trends in the EKE and the role of topographic steering. The manuscript of Lin et al. appears to be a lighter and less complete analysis of the Agulhas Return Current in comparisons to the results published by Fatida et al. (2021) and does not bring new insight into the variability of the Agulhas Return Current.

The method used to identify the path of the Agulhas Return Current also lacked detail. 

Comments on the Quality of English Language

The manuscript was well written and structured. 

Author Response

While the paper is well written, structured and the analysis supports the results, it is not clear to me how the results differ from those published by Fatida et al. (2021). The altimetry data used by Lin et al. is  the same as that used by Fatida et al. (2021). In their paper, Fatida et al. (2021) also report on the same slanting trend of the Agulhas Return Current and point to the wind stress curl as a driving mechanism. The analysis of Fatida et al. (2021) in fact goes further than the results presented here by Lin et al., as Fatida and al. (2021) looks at the seasonal trends and variability in the trends over the period of analysis. The paper of Fatida et al. (2021) also looks at the trends in the EKE and the role of topographic steering. The manuscript of Lin et al. appears to be a lighter and less complete analysis of the Agulhas Return Current in comparisons to the results published by Fatida et al. (2021) and does not bring new insight into the variability of the Agulhas Return Current.

R: Thank you for your comment! Our result is consistent with Fadida et al. (2021) in that the ARC axis exhibits a general slant trend, but we achieved three important progresses beyond their research. (1) We improved the ARC axis definition. Fadida et al. seached the maximum SSH gradient within a fixed area, i.e., 36.125-46.125°S and 20.125-70.325°E. Since the downstream jet is less bathymetric restrained and gets closer to the ACC fronts, multiple maximums and their interference occur frequently in this section (Figure 1). Our method, by restricting the search area along the specific ADT contours, obviously improves the identification of ARC jet axis. Furthermore, we adopt different criteria for the search ranges of upstream and downstream regions in the light of their different kinematic characteristics and circulation background.

(2) By adopting this new method, we reveal that the ARC slanting trend is more dependent on the southward shift of the downstream axis and less on the topographic-steering upstream. It is worth noting that in the headstream, the ARC axis trends manifest as the deformation of its large meanders instead of integral northward migration. This result is totally different to the result of Fadida et al. In addition, Fadida et al suggested the connection between the wind stress curl and the ARC axis in the eastern section solely by linear correlation analysis, lacking the evidence of physical mechanism and process. We go further to reveal the detailed dynamical process by which the wind forcing triggers the slanting of ARC axis.

(3) In the discussion, Fadida et al point out the local EKE change may be related to the ARC axis variability. Still, no robust evidence is provided to support their advocation. In this study, we elaborate the different roles of eddy activity played in the ARC axis change. For the headstream west of 35°E, eddy activity process including shedding, propagation and merging accounts for the ARC axis deformation trend. For the midstream and downstream east of 35°E, however, the ARC axis migration may dominate the local EKE pattern by changing the background circulation and energy cascade direction.

 

The method used to identify the path of the Agulhas Return Current also lacked detail. 

R: More detailed description is added in section 2.3.

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have diligently addressed the questions I raised, resulting in a notable improvement in the paper's quality. Additionally, I recommend dividing the fourth section into a discussion (Section 4) and a conclusion (Section 5) for the article.

Author Response

Thank you for your suggestion! We now divide the discussion and conclusion into two sections in the revised manuscript.