Automated Eddy Identification and Tracking in the Northwest Pacific Based on Conventional Altimeter and SWOT Data

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In the manuscript titled “Automated Eddy Identification and Tracking in the Western Pacific: Assessing SWOT’s Contribution”, the authors evaluated the capability of PET algorithm in identifying oceanic eddies with different spatial scales based on EGF simulations. Through the PET algorithm, they identified mesoscale eddies in the western Pacific and reveal their spatiotemporal characteristics based on nadir-looking altimetry data. The submesoscale eddy-like structures in the SWOT data are also identified and discussed. I think it will be of interest to readers of Remote Sensing. I have some issues that I would like the authors to address before this paper is accepted for publication.

Major:

1. The oceanic eddies mentioned in the manuscript only include mesoscale eddies and submesoscale eddies, and their spatial scales are usually bounded by the Rossby deformation radius. Chelton et al. (2011) divided mesoscale eddies into eddies with lager and smaller spatial scales, but these eddies are all mesoscale eddies. The “Large-scale eddies” and “Small-scale eddies” are wrong terms. Thus, the classification based on spatial scales of eddies in this manuscript is also incorrect and inconsistent with the cited literatures. Please check and modify relevant texts.
2. The Discussion (Lines 590-596) on frequent occurrence of submesoscale ‘pseudo-eddies’ within and around the peripheries of mesoscale eddies is inaccurate. These active submesoscale ‘pseudo-eddies’ are associated with mesoscale strain-induced frontogenesis, which is an important generation mechanism of submesoscale processes.

Minor:

1. The full names corresponding to abbreviations in the abstract should be given. The order of full names and corresponding abbreviations such as SSH and SWOT is confusing in the text.
2. In altimetry data, the ADT is the same as SSH, but they are used simultaneously in the text.
3. The description about two wide swaths of SWOT and the gap between them is inaccurate. Please modify it according to relevant literatures.
4. Line 454: “Mean seasonal cycle of the” --> “The climatologically monthly mean”
5. Figures 9 and 12 both show the eddy trajectories, but we don't know where these eddies generate or disappear.
6. In Figures 17 and 18, the SLA obtained by ascending and descending passes are showed simultaneously in the same panel and they cover each other. The phenomenon may be seen more clearly when plotting SLA from ascending or descending passes only.
7. The SWOT orbits in Figure 19 should be labeled to help readers understand the SWOT observations.

 

Author Response

Assessing SWOT’s Contribution”, the authors evaluated the capability of PET algorithm in identifying oceanic eddies with different spatial scales based on EGF simulations. Through the PET algorithm, they identified mesoscale eddies in the western Pacific and reveal their spatiotemporal characteristics based on nadir-looking altimetry data. The submesoscale eddy-like structures in the SWOT data are also identified and discussed. I think it will be of interest to readers of Remote Sensing. I have some issues that I would like the authors to address before this paper is accepted for publication.

A: Thank you very much for your positive comments. We have carefully revised the manuscript according to your comments and suggestions.

Major:

1. The oceanic eddies mentioned in the manuscript only include mesoscale eddies and submesoscale eddies, and their spatial scales are usually bounded by the Rossby deformation radius. Chelton et al. (2011) divided mesoscale eddies into eddies with lager and smaller spatial scales, but these eddies are all mesoscale eddies. The “Large-scale eddies” and “Small-scale eddies” are wrong terms. Thus, the classification based on spatial scales of eddies in this manuscript is also incorrect and inconsistent with the cited literatures. Please check and modify relevant texts.

A: Yes, the 'small-scale eddies' as said by Chelton et al. (2011) refer to mesoscale eddies at the lower end of the mesoscale spectrum and remain within the mesoscale range. In contrast, the small-scale eddies referred to in this study are eddies with diameters less than 15 km, corresponding to submesoscale eddies, which differ from the "small-scale" eddies described in Chelton et al. (2011). To more clearly delineate the scale range of eddies identified and tracked in this study, we classify oceanic eddies into three categories: large-scale eddies, mesoscale eddies,  and submesoscale. This distinction is clarified in lines 44-60 of the revised manuscript. To avoid confusion, the term "submesoscale eddies" is consistently used throughout the paper instead of "small-scale eddies".

 

2. The Discussion (Lines 590-596) on frequent occurrence of submesoscale ‘pseudo-eddies’ within and around the peripheries of mesoscale eddies is inaccurate. These active submesoscale ‘pseudo-eddies’ are associated with mesoscale strain-induced frontogenesis, which is an important generation mechanism of submesoscale processes.

A: Thank you for your comments. We have replaced the term submesoscale "pseudo-eddies" with submesoscale features, and have clarified that these features are associated with mesoscale strain-induced frontogenesis.

Minor:

1. The full names corresponding to abbreviations in the abstract should be given. The order of full names and corresponding abbreviations such as SSH and SWOT is confusing in the text.

A: Revised.

2. In altimetry data, the ADT is the same as SSH, but they are used simultaneously in the text.

A: Yes, in altimetry data, ADT and SSH are same; however, ADT incorporates the mean dynamic topography (MDT). Many geographically correlated eddies are formed near coasts or bathymetric changes, result from wind-orographic effects or current retroflections, and have an imprint on the MDT. Consequently, differences may arise in eddy detection and tracking when using ADT versus SSH. Pegliasco et al. (2022) pointed out that using ADT instead of SLA maps as input can improve eddy identification, particularly in regions with strong sea surface height (SSH) gradients. In this study, we use SLA (i.e., SSH minus the long-term mean SSH), and this has been applied consistently throughout the manuscript.

3. The description about two wide swaths of SWOT and the gap between them is inaccurate. Please modify it according to relevant literatures.

A: Revised in lines 172-174.

4. Line 454: “Mean seasonal cycle of the” --> “The climatologically monthly mean”

A: Revised.

5. Figures 9 and 12 both show the eddy trajectories, but we don't know where these eddies generate or disappear.

A: In the tracking algorithm, an eddy that does not intersect with another at any given time is considered to have dissipated, signaling the end of the tracking process for that particular eddy. The generation and dissipation regions of the long-lived eddy (lifetime > 32 weeks) shown in Figure 10c are described in lines 509–515.

6. In Figures 17 and 18, the SLA obtained by ascending and descending passes are showed simultaneously in the same panel and they cover each other. The phenomenon may be seen more clearly when plotting SLA from ascending or descending passes only.

A: Revised.

7. The SWOT orbits in Figure 19 should be labeled to help readers understand the SWOT observations.

A: labeled.

 

Reviewer 2 Report

Comments and Suggestions for Authors

Upon reviewing the manuscript, it is evident from Section 5.1 that the authors have conducted meticulous and comprehensive work. However, the connection between these results and the paper’s title remains unclear. A critical issue lies in the lack of logical coherence between Sections 5.1 and 5.2. The authors should either strengthen the linkage and discussion between these sections, consider splitting the content into two separate papers, or revise the title to align with the current scope. Given these structural and thematic inconsistencies, significant revisions or resubmission are recommended to enhance clarity and focus.

Specific Comments:

1. Line 16: The abbreviation "EGF PET" should be spelled out in full upon its first mention.

2. Lines 42–44: The definitions of "large-scale" and "mesoscale eddies" must align with widely accepted standards in the field and remain consistent throughout the manuscript. For example, the description in Line 229 ("Mesoscale eddies are known as oceanic features with diameters ranging from 10 to 200 km") should be harmonized with earlier definitions to avoid ambiguity.

Author Response

Upon reviewing the manuscript, it is evident from Section 5.1 that the authors have conducted meticulous and comprehensive work. However, the connection between these results and the paper’s title remains unclear. A critical issue lies in the lack of logical coherence between Sections 5.1 and 5.2. The authors should either strengthen the linkage and discussion between these sections, consider splitting the content into two separate papers, or revise the title to align with the current scope. Given these structural and thematic inconsistencies, significant revisions or resubmission are recommended to enhance clarity and focus.

A: Thank you very much for your comments and suggestions. The main results of this study are presented in Section 5.1, where mesoscale eddies are identified and tracked using conventional altimeter data. Section 5.2 presents a preliminary attempt to detect submesoscale eddies using SWOT data. A bridging paragraph has been added between the two sections (Lines 601-607). Accordingly, we have revised the title of the manuscript to: "Automated Eddy Identification and Tracking in the Western Pacific Based on Conventional Altimeter and SWOT Data." We hope that the revised title more accurately reflects the scope of the study. Thank you again for your constructive feedback.

Specific Comments:

1. Line 16: The abbreviation "EGF PET" should be spelled out in full upon its first mention.

A: Revised.

2. Lines 42–44: The definitions of "large-scale" and "mesoscale eddies" must align with widely accepted standards in the field and remain consistent throughout the manuscript. For example, the description in Line 229 ("Mesoscale eddies are known as oceanic features with diameters ranging from 10 to 200 km") should be harmonized with earlier definitions to avoid ambiguity.

A: Thank you for your comments. An explanation has been added in Lines 44-60 of the revised manuscript, and the definitions of eddy scales have been made consistent across the entire text.

Reviewer 3 Report

Comments and Suggestions for Authors

This study aimed at identifying large, mesoscale and submesoscale eddies. The conclusions could have something new about submesoscale eddies. However, in the present version, the authors should concentrate on the new findings and display the related figures about the new findings.

• In the abstract, The conclusion “Our analysis revealed strong associations between eddies and major currents, with peak activity around 20°N” was not supported in the present manuscript. Meanwhile, the conclusions “strong associations between eddies and major currents “ and “peak activity around 20°N” are not new for the readers. It is important to provide new findings. To bring the new findings, the authors should include more references, not only the detecting methods but also the findings in previous references.

 

• In the abstract and conclusion, the conclusion “a seasonal peak in spring” was not supported in the present version.  The figures do not provide this information.
• “this study attempts to enhance the precision and reliability of eddy tracking methodologies”. According to the present version, the precision and reliability have not been compared. The authors should provide how to quantify the precision and reliability of eddy tracking methodologies.

Comments on the Quality of English Language

The English Language is good but logical connection is weak.

Author Response

This study aimed at identifying large, mesoscale and submesoscale eddies. The conclusions could have something new about submesoscale eddies. However, in the present version, the authors should concentrate on the new findings and display the related figures about the new findings.

A: Thank you very much for your positive comments and suggestions. Regarding the new findings, although the PET algorithm is relatively mature, there has been no specific study or systematic analysis of eddy identification and tracking over a period exceeding 20 years in the western Pacific. This region is characterized by dynamic oceanic conditions and complex circulation patterns, making it a representative area for investigating the formation and evolution of ocean eddies. Moreover, our results differ in some aspects from those in the META3.1 global eddy dataset within the study region, as discussed in the discussion section. This study could provides reference for the mesoscale eddies study in the western Pacific, like spatiotemporal statistics, generation, and dissipation mechanisms. In addition, we explored the potential of using SWOT satellite data to detect submesoscale eddies, offering new results for the SWOT application community.

In the abstract, The conclusion “Our analysis revealed strong associations between eddies and major currents, with peak activity around 20°N” was not supported in the present manuscript. Meanwhile, the conclusions “strong associations between eddies and major currents “ and “peak activity around 20°N” are not new for the readers. It is important to provide new findings. To bring the new findings, the authors should include more references, not only the detecting methods but also the findings in previous references.

A: The strong associations between eddies and major currents are supported by Figures 7, 8, and 10. Both eddy count and amplitude reach their maximum values along major currents, particularly in the Kuroshio region. As shown in the eddy trajectories in Figure 10, eddies with lifespans longer than 8 weeks exhibit a clear concentration along the Kuroshio path. Moreover, long-lived eddies with durations exceeding 32 weeks are also predominantly located near major currents, highlighting the strong linkage between eddy activity and large-scale oceanic circulation.

The peak eddy activity around 20°N is evident in Figure 7. Grid-based statistics of eddy counts show the highest eddy density occurring between 20°N and 30°N, particularly along the Kuroshio path. We have added and listed relevant previous studies already cited in the manuscript that support these observations, including:

Lines 406-408：This finding is consistent with the research by Qiu (1999), which identified 18°N-25°N as a frequent eddy occurrence zone , and with the strong eddy kinetic energy (EKE) observed in the 18°N-25°N, 125°E-140°E region in the 8-year satellite altimetry reanalysis by Lee et al., (2013).

Lines 512-514: Around 25°N, eddies form within the shear zone between the eastward Subtropical Countercurrent (18°N-25°N) and the westward North Equatorial Current (Qiu et al., 1999). They propagate westward at several centimeters per second, evolve during transit, and eventually dissipate upon encountering the Kuroshio Current (Qiu, 2001).

Line 386-388: These regions are known for their high eddy activity due to the interaction of various ocean currents and the complex topography (Zhang et al., 2017; Shi et al., 2022).

Lines 408-412: The higher eddy count in this latitude range can be attributed to the dynamic interactions between the North Equatorial Current, the Kuroshio, and the regional wind patterns, which create favorable conditions for eddy formation and propagation (Qiu and Chen, 2005).

 

 In the abstract and conclusion, the conclusion “a seasonal peak in spring” was not supported in the present version.  The figures do not provide this information.

A: In Figure 9b, the climatological monthly mean number of cyclonic and anticyclonic eddies from 2000 to 2022 shows a peak in March , which is consistent with the findings of Yin et al., (2019), who reported a notable peak in the number of cyclonic eddies between January and April in this region (Lines 468-470).

 

“this study attempts to enhance the precision and reliability of eddy tracking methodologies”. According to the present version, the precision and reliability have not been compared. The authors should provide how to quantify the precision and reliability of eddy tracking methodologies.

A: Thank you very much for your comments. We have revised this sentence：this study attempts to reconstruct the 22-year spatiotemporal variability of mesoscale eddies in the western Pacific and to evaluate the robustness of eddy tracking methodologies。To assess reliability, we compared the identified eddy trajectories with Surface Velocity Program (SVP) drifter observations (Figure 12).

 

Comments on the Quality of English Language: The English Language is good but logical connection is weak.

A: Thank you very much for your positive comments. We have strengthened the logical connections throughout the text, including the insertion of a bridging paragraph between Sections 5.1 and 5.2, and so on.

Reviewer 4 Report

Comments and Suggestions for Authors

The authors utilized the Py_Eddy_Tracking code to identify and track eddies in the western Pacific. They also incorporated the new SWOT dataset in their eddy detection process. However, the overall novelty of the present paper is limited. The statistics of eddies in the western Pacific are largely consistent with previous studies. I recommend that the authors further explore the applications of the SWOT dataset in more depth.

Author Response

The authors utilized the Py_Eddy_Tracking code to identify and track eddies in the western Pacific. They also incorporated the new SWOT dataset in their eddy detection process. However, the overall novelty of the present paper is limited. The statistics of eddies in the western Pacific are largely consistent with previous studies. I recommend that the authors further explore the applications of the SWOT dataset in more depth.

A: Thank you very much for your positive comments. The novelty of this study lies in the long-term reconstruction of mesoscale eddy distributions and trajectories over the western Pacific using conventional altimeter data spanning more than two decades, a region and timescale that have not been systematically analyzed used by PET method in previous studies. This study could provides reference for the mesoscale eddies study in the western Pacific, like spatiotemporal statistics, generation, and dissipation mechanisms.

We explore the potential of SWOT satellite data for detecting submesoscale eddies, presenting one of the first attempts to identify such features using high-resolution SWOT observations. At present, there are still challenges in using SWOT data to identify submesoscale eddies. It remains difficult to ensure that the detected small-scale features below 50 km are indeed real signals rather than artifacts of orbital noise (Tom Farrar et al., 2024 SWOT Science Meeting). Further evidence is needed to validate these detections, and more in-depth investigations will be carried out in our future work.

Tom Farrar, Rosemary Morrow, Francesco d’Ovidio., Ocean view on 3 main questions. 2024 SWOT Science Meeting.

Reviewer 5 Report

Comments and Suggestions for Authors

This paper is about testing an eddy detection and tracking algorithm, which was originally developed by Mason et al.,2014, and appling it to altimetry sea level datasets in the western pacific region. The authors use more than 20 years of sea level anomaly datasets (from 2002 to 2022) to track the eddies in different size and scales. The algorithm is then applied to SWOT swath sea level measurements. There is no solid conclusion from this study. The arguments and dynamic exploration of the eddy evolution and distribution are vague and plausible. How are the eddies interacted with the mean flow (Kuroshio),climate, and other factors is not well discussed. The paper is lengthy and not well organized to reach to a clear point showing what the research goals are and what the study significance is. Thus I suggest the authors make a thorough revision to bring the research goals and significance up by demonstrating the significant contribution of this research to the general  oceanography science. Thus, this paper should not be published until being thoroughly revised.

 

 

Major comments:

1. The long term time series of 20 years eddy distribution should yield some or correlate with regional climate signals. Clearly, the authors did not do any further analysis on this and lose the argument why they are looking at the 20 years of the data sets.

2. The monthly variation should be further explored with regional wind patterns or ocean flow variations, rather than a vague explanation of wind pattern changes.

3. Looking at the EKE and radii of an eddy is an important part to characterizing the eddy itself. The overall EKE from the mean flow and turbulent fields should be also looked so that they can be quantified to correlate with eddy activities (eddy numbers, regions, amplitude etc).

4. This region has strong internal tide/wave signals, which can not be completely removed in SLA corrections, especially for incoherent internal tides displayed on the ocean surface. The authors should make efforts to discuss their impact. Is there a way to remove them or reduce their impacts? High pass filters may not work well, especially for SWOT images, near Luzon strait and Sulawesi sea area.

5. The SWOT has total swath of 120km, certainly the contour tracking will break for large eddies. This should be remedied when looking at the datasets from daily cal/val orbit as improvements of PET methods. For stitched images at the science phase of SWOT, did the author consider the time difference of different tracks in tracking the eddies, especially small scale eddies?

6. The paper is very lengthy. For example, since PET is a matured, validated method being used by many researchers, the Section 4 can be much simplified.

    

Minor comments:

The general writing of this paper looks good with clear language. Some sentences miss some wording and need some grammar/typo check.

 

Figure 4, c) Are the ER and SR the same?

Figure 12. The drifter should be label at the starting point so the readers know the travel direction of the eddies.

Author Response

This paper is about testing an eddy detection and tracking algorithm, which was originally developed by Mason et al.,2014, and appling it to altimetry sea level datasets in the western pacific region. The authors use more than 20 years of sea level anomaly datasets (from 2002 to 2022) to track the eddies in different size and scales. The algorithm is then applied to SWOT swath sea level measurements. There is no solid conclusion from this study. The arguments and dynamic exploration of the eddy evolution and distribution are vague and plausible. How are the eddies interacted with the mean flow (Kuroshio),climate, and other factors is not well discussed. The paper is lengthy and not well organized to reach to a clear point showing what the research goals are and what the study significance is. Thus I suggest the authors make a thorough revision to bring the research goals and significance up by demonstrating the significant contribution of this research to the general oceanography science. Thus, this paper should not be published until being thoroughly revised.

A: Thank you very much for your comments. Please refer to the Major Comments section for detailed responses.

Major comments:

1. The long term time series of 20 years eddy distribution should yield some or correlate with regional climate signals. Clearly, the authors did not do any further analysis on this and lose the argument why they are looking at the 20 years of the data sets.

A: We thank the reviewer for this valuable suggestion. In our revised manuscript, we present the monthly climatology of eddy counts (Figure 9b) and associate the observed seasonal variability with known large-scale climate drivers, particularly ENSO phases (Figure 9a and Lines 443-456). ENSO events modulate both wind and current patterns in the western Pacific, and their linkage to eddy activity is well established in previous studies (e.g., Simanungkalit et al., 2018). By showing the statistical alignment between eddy activity and ENSO, we indirectly account for seasonal variations in wind and flow without introducing additional datasets. We believe this approach provides a clearer, more integrated explanation while maintaining focus on the core objectives of this study.

2. The monthly variation should be further explored with regional wind patterns or ocean flow variations, rather than a vague explanation of wind pattern changes.

A: We appreciate the suggestion to include wind field data. However, Figure 9 already captures seasonal and interannual eddy variability through robust statistical links with ENSO phases. ENSO events inherently reflect large-scale wind anomalies in the western Pacific. Figure 9a shows that peaks in eddy counts often coincide with La Niña events, while El Niño periods align with troughs. Figure 9b reveals clear seasonal cycles in eddy occurrence.

Given this strong ENSO–eddy correlation, which subsumes major wind-driven variability, we believe additional wind field data would add complexity without significantly improving the interpretation. We thus focus on ENSO as a comprehensive climatic driver.

3. Looking at the EKE and radii of an eddy is an important part to characterizing the eddy itself. The overall EKE from the mean flow and turbulent fields should be also looked so that they can be quantified to correlate with eddy activities (eddy numbers, regions, amplitude etc).

A: We agree that EKE is a critical parameter for characterizing eddy dynamics. In our study, we have already quantified and analyzed eddy kinetic energy (EKE) based on geostrophic velocities derived from SLA (Figure 6c–d), and we correlated EKE with eddy radius, amplitude, and spatial distribution. This provides meaningful insight into the intensity and structure of eddies across the region. While a decomposition of EKE into mean and turbulent components could further enrich the analysis, it requires longer-term velocity field data or model output beyond the scope and objectives of this observational study. We believe the current EKE analysis sufficiently supports the link between eddy strength and spatial patterns, particularly in high-energy regions like the Kuroshio and northern SCS.

4. This region has strong internal tide/wave signals, which can not be completely removed in SLA corrections, especially for incoherent internal tides displayed on the ocean surface. The authors should make efforts to discuss their impact. Is there a way to remove them or reduce their impacts? High pass filters may not work well, especially for SWOT images, near Luzon strait and Sulawesi sea area.

A: We appreciate the reviewer’s observation regarding internal tide and wave signals. Indeed, incoherent internal tides are difficult to remove entirely, particularly in complex regions like the Luzon Strait. Our SLA dataset applies state-of-the-art corrections, including the FES22 ocean tide model and internal tide filtering in the SWOT L3 product. While high-pass filtering may not fully eliminate internal tide contamination, it helps isolate mesoscale and submesoscale features relevant to this study. We acknowledge this limitation in Section 6.1 and agree that future work incorporating more advanced tide-removal methods or internal wave models could further reduce uncertainties in eddy detection.

5. The SWOT has total swath of 120km, certainly the contour tracking will break for large eddies. This should be remedied when looking at the datasets from daily cal/val orbit as improvements of PET methods. For stitched images at the science phase of SWOT, did the author consider the time difference of different tracks in tracking the eddies, especially small scale eddies?

A: We acknowledge the reviewer’s concern regarding the limitations of SWOT’s swath coverage. As noted, the PET algorithm relies on closed SLA contours to identify eddies. For mesoscale eddies, the 120 km SWOT swath may truncate large structures, especially near swath edges, preventing the identification of complete closed contours. As a result, mesoscale eddies cannot always be reliably extracted from SWOT’s science-phase data alone.

However, during SWOT's Cal/Val phase, the 1-day repeat orbit provides temporally dense observations that preserve eddy continuity within the same swath. In our study, we leveraged these high-temporal-resolution Cal/Val datasets to better capture submesoscale structures and mitigate contour fragmentation (Section 5.2.1, Figure 14–15). For the science orbit, we analyzed stitched images with awareness of the temporal gaps between adjacent tracks. Because SWOT samples different longitudes at different times, we did not attempt time-continuous tracking across swaths. Instead, we treated these as spatial snapshots, focusing on characterizing submesoscale features rather than tracking eddy evolution in time. We have clarified these limitations and the rationale for our approach in Section 6.2.

6. The paper is very lengthy. For example, since PET is a matured, validated method being used by many researchers, the Section 4 can be much simplified.

A: We have streamlined Section 4 by shortening the explanation of the PET algorithm and removing redundant details about contour constraints and eddy tracking procedures. While PET is a well-established method, we retained a concise description and simulation-based validation to clarify how it performs under different eddy scales, particularly submesoscale features relevant to SWOT. These updates aim to balance methodological clarity with manuscript conciseness.

Minor comments:

The general writing of this paper looks good with clear language. Some sentences miss some wording and need some grammar/typo check. 

A: Thank you very much for your positive comments. We have revised the grammar and corrected typos throughout the manuscript.

Figure 4, c) Are the ER and SR the same?

A: While the ER (Effective Radius) and SR (Speed Radius) appear similar in Figure 4(c), they are not exactly the same. This is clarified in Section 4.2, where we explain their definitions and characteristics. For small-scale or submesoscale eddies, the relative difference between ER and SR tends to decrease due to the compact structure and limited radial velocity gradient. However, even minor differences can represent meaningful variation at small scales. We have added a clarifying note in Section 4.2 (Lines 267-269) and also refer to the existing content in Lines 655-662  to highlight this point.

Figure 12. The drifter should be label at the starting point so the readers know the travel direction of the eddies.

A: We have updated Figure 12 by adding markers at the starting points of the drifter trajectories. This revision clarifies the direction of drifter movement and improves the visualization of eddy travel paths.

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have well addressed my previous comments.  I think it can be accepted in the present form. 

Author Response

The authors have well addressed my previous comments.  I think it can be accepted in the present form. 

A: Thank you very much for your positive comments.

Reviewer 2 Report

Comments and Suggestions for Authors

detailed comments：
Line 2  "Western Pacific" should be "western North Pacific" according the study area. Please check the full text to ensure that similar issues have been revised.  
Line 17 Surface Velocity Program (SVP) 
Line 24-25 “ with a concentration east of Taiwan”revise to“mainly concentrated east of Taiwan Island”
Line 138  and Line 139:    Taiwan Island, and Please check the full text to ensure that similar issues have been revised.  
Line 179  Figure 1，Clarify data sources for bathymetric mapping in the caption
Line 261 Figure 2. revise the caption as "Workflow for eddy identification with PET".
Figure 4: The size reference in panel (c) is ambiguous. We recommend clearly delineating the spatial extent of panel (c) within panel (b) to facilitate a clearer visual comparison of eddy scales across panels (a)-(c).
Figure 5: The units for the color-shaded parameter (height) in panel (a) are not specified. Please clarify this information.
Line 637: In the subtropical Pacific, many mesoscale eddies originate outside the study area and propagate westward into the research domain. The methodology for calculating their lifespan remains unclear: Is the lifespan measured from the date of their first entry into the study area, or from their initial formation date (outside the study area)? The manuscript should explicitly address this distinction to ensure methodological transparency.
Line 999：either "smaller" or "1/4°" is not true, please check
Line 999-1000: the "pixels" here are different? Add detailed information.

Author Response

Line 2  "Western Pacific" should be "western North Pacific" according the study area. Please check the full text to ensure that similar issues have been revised.  

A: Revised to "Northwest Pacific".
Line 17 Surface Velocity Program (SVP) 

A: Revised.
Line 24-25 “ with a concentration east of Taiwan”revise to“mainly concentrated east of Taiwan Island”

A: Revised.
Line 138  and Line 139:    Taiwan Island, and Please check the full text to ensure that similar issues have been revised.  

A: Revised.
Line 179  Figure 1，Clarify data sources for bathymetric mapping in the caption

A: Revised.
Line 261 Figure 2. revise the caption as "Workflow for eddy identification with PET".

A: Revised.
Figure 4: The size reference in panel (c) is ambiguous. We recommend clearly delineating the spatial extent of panel (c) within panel (b) to facilitate a clearer visual comparison of eddy scales across panels (a)-(c).

A: The effective radii of the eddies in (a) to (c) are 113.89 km, 47.73 km, and 6.02 km, respectively (see Table 2).
Figure 5: The units for the color-shaded parameter (height) in panel (a) are not specified. Please clarify this information.

A: Figure 5. (a) SLA on May 1, 2023 by applying a 500 km high-pass filter, unit is m.
Line 637: In the subtropical Pacific, many mesoscale eddies originate outside the study area and propagate westward into the research domain. The methodology for calculating their lifespan remains unclear: Is the lifespan measured from the date of their first entry into the study area, or from their initial formation date (outside the study area)? The manuscript should explicitly address this distinction to ensure methodological transparency.

A: A clarification has been added in Line 537: "Here, the eddy lifetime refers to the duration of the eddy trajectories within the study area."

Line 999：either "smaller" or "1/4°" is not true, please check

A: We agree that the description was inaccurate, and therefore, it has been removed.
Line 999-1000: the "pixels" here are different? Add detailed information.

A：We have add detailed in Lines 826-828: For submesoscale eddies, which are generally smaller than 10 km in radius, an excessive number of pixels may fail to accurately capture their compact structure and can lead to an overestimation of their effective radius (ER) during detection.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript has been improved greatly.  I am not very satisfied the responses. Some important previous studies were not mentioned and this will cause some results not compared with other studies. Thus, some of the conclusions are not new.  Adding some discussions is necessary. This is important for future studies and progress. 

1) The finding "Our analysis revealed strong associations between eddies and major currents, with peak activity around 20°N and along the Kuroshio Current and northern South China Sea." is not new. Many studies have found this. If the authors want to add this in the abstract, Please make a new statement. 

2) The methods compared with others is necessary, such as:

Ioannou, A.; Guez, L.; Laxenaire, R.; Speich, S. Global Assessment of Mesoscale Eddies with TOEddies: Comparison Between Multiple Datasets and Colocation with In Situ Measurements. Remote Sens. 2024, 16, 4336. https://doi.org/10.3390/rs16224336

Meng et al. Oceanic mesoscale eddy in the Kuroshio extension: Comparison of four datasets 

Atmospheric and Oceanic Science Letters, https://doi.org/10.1016/j.aosl.2020.100011

.....

Author Response

The manuscript has been improved greatly.  I am not very satisfied the responses. Some important previous studies were not mentioned and this will cause some results not compared with other studies. Thus, some of the conclusions are not new.  Adding some discussions is necessary. This is important for future studies and progress. 

A：Thank you very much for your positive comments. We have add some discussions according to your comments.

• The finding "Our analysis revealed strong associations between eddies and major currents, with peak activity around 20°N and along the Kuroshio Current and northern South China Sea." is not new. Many studies have found this. If the authors want to add this in the abstract, Please make a new statement. 

A：We have deleted this sentence.

2) The methods compared with others is necessary, such as:

Ioannou, A.; Guez, L.; Laxenaire, R.; Speich, S. Global Assessment of Mesoscale Eddies with TOEddies: Comparison Between Multiple Datasets and Colocation with In Situ Measurements. Remote Sens. 2024, 16, 4336. https://doi.org/10.3390/rs16224336

Meng et al. Oceanic mesoscale eddy in the Kuroshio extension: Comparison of four datasets Atmospheric and Oceanic Science Letters, https://doi.org/10.1016/j.aosl.2020.100011

A: The characteristics of the TOEddies method and its comparison with META3.2 have been included in the discussion (Lines 750-753).

Reviewer 4 Report

Comments and Suggestions for Authors

The quality of this paper has been much improved.

Author Response

The quality of this paper has been much improved.

A：Thank you very much for your your positive feedback and comments.

 

Reviewer 5 Report

Comments and Suggestions for Authors

This paper has been revised as suggested. I thank the authors quick response to address my questions and concerns. While there are still rooms for improvements to elaborate further the inter-annual oceanic and atmospheric variability controlling the eddy activities, I think this paper can be published to Remote Sensing as current shape with minor revision.

Minor comments: 

The authors added a new reference  at line 560, but it is not shown in the reference list. Please add. Lee et al., 2013.

Author Response

This paper has been revised as suggested. I thank the authors quick response to address my questions and concerns. While there are still rooms for improvements to elaborate further the inter-annual oceanic and atmospheric variability controlling the eddy activities, I think this paper can be published to Remote Sensing as current shape with minor revision.

A：Thank you very much for your your positive comments.

Minor comments: 

The authors added a new reference  at line 560, but it is not shown in the reference list. Please add. Lee et al., 2013.

A: added.