Long-Term Variation in Mesoscale Eddy Activity Around the Kuroshio in the East China Sea During 1993–2023
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis is a solid study presenting valuable long-term trend analysis of mesoscale eddy activity in the ECS. The finding of opposing trends for cyclonic and anticyclonic eddies is particularly interesting. However, the manuscript would significantly benefit from addressing the following points.
1. To assess the sensitivity of the key conclusion linking axis shift to eddy trends, a comparative analysis using alternative, common definitions (the 1 m/s isotach, the SSH front, ...) is strongly recommended. Discussing whether the inferred northwestward shift is consistent across methods would greatly enhance confidence in this mechanistic link.
2. Significant EKE increase without a concurrent rise in eddy count? It needs a more detailed diagnostic. Are eddies becoming more long-lived? Is the increasing EKE manifested in stronger or larger individual eddies? Are there trends in eddy propagation speed, merger/split events, or trapping capability? A more comprehensive eddy census is crucial to trace where the added energy is going.
3. The 28-day lifetime cutoff excludes a potentially dynamic part of the eddy field. To test the hypothesis that the "missing" EKE trend resides in shorter-lived features, it might be useful to analyze the population distribution of eddies with lifetimes below the 28-day threshold. Has the proportion or energy of short-lived eddies changed? This is a key validation of the detection methodology's impact on the conclusions.
4. Increased energy injection could also fuel submesoscale dynamics. While possibly beyond this paper's scope, briefly discussing this possibility would provide a more complete picture of the kinetic energy cascade and frame the findings within the broader scale interaction context.
5. The linear trend analysis may be conflated with low-frequency oscillations (PDO). Employing methods like Ensemble Empirical Mode Decomposition to separate timescales? Correlating EKE/eddy property anomalies with climate indices to quantify their influence?
Author Response
Point-to-Point Responses to the Reviewers' Comments
Response to Reviewer #1:
This is a solid study presenting valuable long-term trend analysis of mesoscale eddy activity in the ECS. The finding of opposing trends for cyclonic and anticyclonic eddies is particularly interesting. However, the manuscript would significantly benefit from addressing the following points.
Response: Thank you for your meticulous review and highly constructive feedback on our manuscript. We sincerely appreciate the time and expertise you have dedicated to this work. Your insightful comments have been instrumental in guiding us to substantially improve the clarity, scientific rigor, and overall quality of the manuscript. A detailed, point-by-point response to your comments is provided below.
- To assess the sensitivity of the key conclusion linking axis shift to eddy trends, a comparative analysis using alternative, common definitions(the 1 m/s isotach, the SSH front, ...) is strongly recommended. Discussing whether the inferred northwestward shift is consistent across methods would greatly enhance confidence in this mechanistic link.
Response: We fully agree with your valuable suggestion that a comparative analysis of the Kuroshio Current axis shift constitutes a critical step to enhance the credibility of our conclusion. Below are the findings from the relevant analyses we have carried out.
We followed the approach proposed by prior studies (e.g., Qiu and Chen, 2010; Zhou et al., 2021; Kawakami et al., 2023) to define the Kuroshio Current axis and selected the representative 100 cm sea surface height (SSH) contour as the location of the Kuroshio Current axis. The rationale for selecting the typical SSH contour level was justified by Qiu and Chen (2010), who demonstrated that the 170 cm SSH contour in their study was consistently located in close proximity to the maximum zonal geostrophic velocity, thereby serving as a reliable indicator of the Kuroshio Extension axis. Their SSH dataset was derived by combining the sea level anomaly from the AVISO dataset with the mean dynamic topography presented by Teague et al. (1990). Owing to different reference levels, the 170 cm SSH contour in their study corresponds to the 100 cm absolute dynamic topography (ADT) contour in ours. As illustrated in Figure R1, the 100 cm ADT contour lies near the maxima of , which serves as a good indicator of the Kuroshio Current axis.
Figure R1. Spatial distribution of (a) annual mean and (b-e) seasonal mean absolute dynamical topography (ADT, units: cm) in the East China Sea during 1993~2023. The grey line denotes the 50 m isobath, and the thick black lines denote the 100 cm contour of ADT. The focal study area is indicated by the green diagonal box in both panels. The vectors in (a) are the annual mean geostrophic currents (units: m/s) during 1993~2023. Here, only current speeds exceeding 0.25 m/s are displayed for clarity.
Figure R2 displays the time series of the Kuroshio Extension axis within the study domain, where the axis is defined by the 100 cm ADT contour. The results clearly illustrate that the Kuroshio Extension axis in the East China Sea underwent a pronounced westward (-0.08°/decade) and northward displacement (0.08°/decade) during 1993~2023, and both trends were statistically significant at the 95% confidence level. The Kuroshio axis represented by the 100 cm ADT contour shifted westward in all four seasons, with a westward shift magnitude exceeding 0.10 °/decade in summer, autumn and winter. Meridionally, the Kuroshio axis also showed a northward shift of about 0.06°~0.11°/decade in four seasons. These results are consistent with the Kuroshio displacement patterns identified in our main manuscript (Figure 7).
Additionally, existing literature indicates that the shift of the intense western boundary current axis aligns with findings reported in previous studies (e.g., Yang et al., 2020; Zhou et al., 2025). For instance, as noted by Zhou et al. (2025), "Zonal centroids of both global mesoscale eddy kinetic energy … have shifted shoreward". These results conclusively verify a pronounced northwestward shift of the Kuroshio Current axis within the study area, a dynamic process that likely contributed to the anticyclonic polarity transition of eddies in this area over the satellite altimetry era.
Figure R2. Same as Figure 7 of the manuscript, but with the Kuroshio axis position defined by the 100 cm absolute dynamic topography contour.
Corresponding descriptions have been added to the revised manuscript to elaborate on the relevant physical implications, which further strengthen confidence in the proposed mechanistic link, as you suggested. Please refer to the revised manuscript and Supporting Information for details.
- Significant EKE increase without a concurrent rise in eddycount? It needs a more detailed diagnostic. Are eddies becoming more long-lived? Is the increasing EKE manifested in stronger or larger individual eddies? Are there trends in eddy propagation speed, merger/split events, or trapping capability? A more comprehensive eddy census is crucial to trace where the added energy is going.
Response: Thank you for raising this key scientific issue. These questions are essential to revealing the long-term features of mesoscale eddies in the study area, and we have conducted a comprehensive quantitative eddy census to diagnose the detailed patterns of mesoscale eddies, with all relevant results supplemented in the revised manuscript and Supporting Information.
Q1: Significant EKE increase without a concurrent rise in eddy count?
Yes, our detailed analyses confirm that the total eddy count does not show a concurrent increasing trend with EKE. As shown in Figure R3, the number of cyclonic and anticyclonic eddies exhibits a decreasing and an increasing trend, respectively, with most of these trends being statistically insignificant. This indicates that the pronounced EKE increase is not mainly manifested in the trends in eddy count, but instead is dominated by the long-term variations in the intrinsic properties of mesoscale eddies (i.e., eddy lifetime and amplitude), as elaborated below.
Figure R3. Temporal evolutions of the annual and seasonal sum of eddy occurrence number of (a) cyclonic and (b) anticyclonic eddies during 1993~2023. Linear fitting lines of the time series are plotted with solid lines, and the trends (units: /decade) are also noted, and associated p values in parentheses; statistical significance at the 95% confidence level is denoted by p < 0.05.
Q2: Are eddies becoming more long-lived?
Yes, the lifetime of mesoscale eddies also demonstrated long-term trends, with distinct polarity-dependent patterns. Specifically, cyclonic eddies showed a shortening trend in lifetime over the study period, whereas anticyclonic eddies displayed an increasing trend, as illustrated in Figure R4. The prolonged lifetime of anticyclonic eddies may serve as a key factor contributing to the EKE accumulation in the study domain, given that longer-lived eddies possess a greater capacity to retain and amplify energy. Notably, large magnitudes of the linear trends in eddy lifetime within the study domain are mainly concentrated in the northern meander region. However, the vast majority of trends fail to pass the statistical significance test at the 95% confidence level.
Figure R4. Spatial distributions of (a) annual mean eddy lifetime (EL, units: days) and (b) linear trends (units: days/decade) of total, cyclonic, and anticyclonic eddies in the East China Sea during 1993~2023. The grey line denotes the 50 m isobath. The focal study area is indicated by the green diagonal box in both panels.
Q3: Is the increasing EKE manifested in stronger or larger individual eddies?
The observed increase in EKE is primarily manifested in the enhanced intensity of anticyclonic eddies. Our analysis shows that the amplitude of anticyclonic eddies exhibits an increasing trend, while that of cyclonic eddies shows a decreasing trend (Figure R5). As amplitude represents a key metric characterizing the kinetic intensity of coherent mesoscale eddies, the amplified amplitude of anticyclonic eddies likely acts as the dominant manifestation of the overall EKE increase in the study domain. However, based on the monthly mean time series of eddy amplitude and number (Figure R3), this increase in amplitude does not occur in isolation.
Figure R5. Same as Figure R3, but with the eddy occurrence amplitude (units: cm).
Q4: Are there trends in eddy propagation speed, merger/split events, or trapping capability?
We also characterized the propagation features of coherent mesoscale eddies in the study area (Figure R6). A polarity-based comparison shows that anticyclonic eddies propagate over longer distances than cyclonic eddies, with none exceeding 60 km. This suggests that mesoscale eddies active in the East China Sea generally do not deviate markedly from the current axis, which aligns with the eddy propagation patterns in this region documented in previous studies (e.g., Li et al., 2020; Zhang et al., 2022). Furthermore, both eddy propagation distance and speed (Figure R7) display a polarity-dependent trend: cyclonic eddies show shorter propagation distances and slower speeds, whereas the opposite holds for anticyclonic eddies. Nevertheless, none of these trends achieved statistical significance at the 95% confidence level.
Figure R6. Same as Figure R4, but for the propagation distance during eddy lifetime (units: km).
Figure R7. Same as Figure R4, but for the propagation speed during eddy lifetime (units: km/days).
Trend analyses of eddy merger/split events and their trapping capability were not performed in the present study, as they may lie beyond the scope of this work. Such analyses will be a key focus of our future research. This line of investigation will, on the one hand, further clarify the disparities in the long-term trends of cyclonic and anticyclonic eddies, and on the other hand, allow for an in-depth elucidation of the oceanographic dynamic effects and potential impacts induced by such disparities. We have therefore acknowledged this aspect as one of the limitations of the present study. We sincerely appreciate your valuable comments, which offer important guidance for our future research.
To summarize, despite the absence of a concurrent increase in the total eddy count, the increase in EKE in the study region is primarily manifested in anticyclonic eddies, and is contributed to by the combined effects of their prolonged lifetime and amplified amplitude. Detailed quantitative results from the eddy census are provided in the Supporting Information for further reference. Please refer to the revised manuscript and Supporting Information for details.
- The 28-day lifetime cutoff excludes a potentially dynamic part of the eddy field. To test the hypothesis that the "missing" EKE trend resides in shorter-lived features, it might be useful to analyze the population distribution of eddies with lifetimes below the 28-day threshold. Has the proportion or energy of short-lived eddies changed? This is a key validation of the detection methodology's impact on the conclusions.
Response: Thank you for this critical and insightful suggestion, which provides a key validation of whether and how eddy identification algorithms influence our study’s conclusions and effectively addresses the potential omission of the dynamic short-lived eddy component resulting from the 28-day lifetime threshold.
In the initial eddy identification and tracking process, we only retained eddies with a lifespan of more than 28 days. To test the hypothesis that the "missing" EKE trend may be associated with shorter-lived eddies, we performed targeted supplementary analyses using the Mesoscale Eddy Trajectory Atlas Product (the delayed-time version 3.2 operational product, META3.2; Mason et al., 2014; Pegliasco et al., 2022). Two points should be noted. First, the eddy detection and tracking algorithm adopted in this dataset differs from that used in our study (particularly in specific parameter settings), leading to certain discrepancies in the derived results. Second, the temporal coverage of this dataset ranges from January 1, 1993, to February 9, 2022, which also results in differences in the identified trend patterns. Despite these differences, this dataset includes all eddies with lifetimes longer than 10 days and, as a widely adopted benchmark dataset, can be used to independently verify the reliability of our conclusions.
Statistical results of the META3.2 dataset (Table R1) show that within the study area, short-lived eddies (with a lifespan of 10–28 days) account for 32.1% and 24.1% for cyclonic and anticyclonic eddies of the total population, indicating that approximately one-quarter of all eddies in the East China Sea are short-lived.
Table R1: Census of short-lived (lifetime between 10-28 days) and long-lived (lifetime longer than 28 days) eddies in the study area during 1993-2021 obtained from Mesoscale Eddy Trajectory Atlas Product version 3.2. The proportions of the two eddy types relative to the total number are indicated in parentheses.
|
Number of Eddies |
Short-lived Eddies |
Long-lived Eddies |
Total Eddies |
|
Cyclonic Eddies |
9128 (32.1%) |
19309 (67.9%) |
28437 |
|
Anticyclonic Eddies |
3896 (24.1%) |
12293 (75.9%) |
16189 |
As shown in Figure R8, the ratios of short-lived cyclonic and anticyclonic eddies relative to the total number of the corresponding eddy type are shown. During the period 1993~2021, the ratio of short-lived cyclonic eddies exhibited an increasing trend, whereas that of short-lived anticyclonic eddies showed a decreasing trend. The increasing trend in the ratio of short-lived cyclonic eddies was primarily evident in winter, while the decreasing ratio of short-lived anticyclonic eddies was mainly observed in autumn. This aligns with our hypothesis that winter hydrodynamic conditions are more favorable for the generation of cold eddies, whereas warmer seawater in autumn facilitates the formation of warm eddies. Regrettably, none of the trend estimates passed the statistical significance test at the 95% confidence level. When the criterion was relaxed to the 90% confidence level, however, some of these trends achieved marginal statistical significance (figure not shown).
These supplementary results further confirm that the ratio of long-lived cyclonic eddies derived from the META3.2 product decreased over 1993~2021, while that of long-lived anticyclonic eddies increased—a finding consistent with the conclusions presented in our manuscript. This observation uncovers an interesting pattern: short-lived eddies (lifetime between 10-28 days) in the study area tend to be cyclonic in polarity, whereas long-lived eddies (lifetime longer than 28 days) are inclined to be anticyclonic over the satellite altimetry era. As for the underlying cause of this polarity distinction, we hypothesize that it may be linked to the different eddy generation mechanisms. Longer-lived, larger eddies are produced by the development and pinch-off of meanders in the Kuroshio path, while shorter-lived, smaller eddies are driven by horizontal shear instability (Ji et al., 2018). This warrants further in-depth investigation in our subsequent research.
Figure R8. Same as Figure R3, but with the ratio of short-lived eddies (units: %).
All detailed results from the supplementary analyses on short-lived eddies, including population distribution and proportion trends, are compiled and presented in the Supporting Information for your reference. Please refer to the revised manuscript and Supporting Information for details.
- Increased energy injection could also fuel submesoscale dynamics. While possibly beyond this paper's scope, briefly discussing this possibility would provide a more complete picture of the kinetic energy cascade and frame the findings within the broader scale interaction context.
Response: Thank you for this insightful suggestion, which helps to frame our findings within a broader context of multi-scale ocean dynamic interactions and provides a more comprehensive perspective on the kinetic energy cascade in the study domain. Existing studies (e.g., Ji et al., 2021; Chen et al., 2023) have verified the presence of submesoscale motions in the East China Sea. We fully acknowledge that increased energy injection into the regional ocean system could potentially fuel submesoscale dynamics.
However, it should be noted that an in-depth investigation into submesoscale dynamics and its linkages with mesoscale processes and energy injection relies on high-resolution observational or modeling data. We therefore suggest that future work combine high-resolution numerical simulations to explore this topic, which would allow for a clear depiction of fine-scale submesoscale structures and a quantitative understanding of energy exchange and cascade processes between submesoscale and mesoscale systems.
Correspondingly, we have supplemented relevant discussions on this possibility in the Discussion section of the revised manuscript. Please refer to the revised manuscript for details.
- The linear trend analysis may be conflated with low-frequency oscillations (PDO). Employing methods like Ensemble Empirical Mode Decomposition to separate timescales? Correlating EKE/eddy property anomalies with climate indices to quantify their influence?
Response: Thank you for this insightful and critical suggestion, which is essential for disentangling the influence of low-frequency oscillations from the linear trend signals and thus improving the robustness of our trend analysis results. In response to your questions, we have conducted targeted supplementary analyses as recommended.
We adopted the Ensemble Empirical Mode Decomposition (EEMD) algorithm to perform signal decomposition on the EKE anomaly time series (ensemble size=500 and variance of white noise=0.2), which yields three Intrinsic Mode Functions (IMFs) and the nonlinear trend component represented by the residual term. Results illustrated in Figure R9 reveal a distinct nonlinear increasing trend in EKE, with this upward tendency consistent across all seasonal phases. Notable disparities exist among the seasonal trends. This also indicates that an increasing trend in EKE over the study does exist after removing the low-frequency signals. Naturally, there exist some quantitative differences compared with the trend we previously obtained using the least squares method.
Figure R9. Same as Figure 2 of the manuscript, but with the nonlinear trends derived from the Ensemble Empirical Mode Decomposition algorithm (grey dashed lines).
In addition, quantitative analyses indicate that the PDO index (Pacific Decadal Oscillation, https://www.data.jma.go.jp/tcc/tcc/products/elnino/decadal/pdo_doc.html) has no significant correlation with regional EKE, and thus may not be the dominant factor driving the long-term variations of mesoscale eddies in this region. This, in fact, constitutes a key focus of our follow-up research.
In the present study, we primarily adopted the least squares method for the extraction of linear trends—a simple and efficient algorithm for extracting trend signals. However, ocean dynamics is inherently nonlinear, meaning the derived linear trends may not fully capture the actual dynamical characteristics of the study area. Accordingly, we plan to employ the EEMD algorithm to analyze the nonlinear trend characteristics of EKE and eddy properties in subsequent research. We appreciate your enlightening suggestion, which we trust will contribute to enabling a more accurate and comprehensive characterization of the variations in eddy features in the Kuroshio region of the East China Sea.
Correspondingly, we have supplemented relevant discussions on this possibility in the Discussion section of the revised manuscript. Please refer to the revised manuscript for details.
References:
Chen, J.; Zhu, X.-H.; Zheng, H.; Wang, M. Submesoscale Dynamics Accompanying the Kuroshio in the East China Sea. Front. Mar. Sci. 2023, 9, 1124457, doi:10.3389/fmars.2022.1124457.
Ji, J.; Dong, C.; Zhang, B.; Liu, Y.; Zou, B.; King, G.P.; Xu, G.; Chen, D. Oceanic Eddy Characteristics and Generation Mechanisms in the Kuroshio Extension Region. J. Geophys. Res. Oceans 2018, 123, 8548–8567, doi:10.1029/2018JC014196.
Ji, Y.; Xu, G.; Dong, C.; Yang, J.; Xia, C. Submesoscale Eddies in the East China Sea Detected from SAR Images. Acta Oceanol. Sin. 2021, 40, 18–26, doi:10.1007/s13131-021-1714-5.
Kawakami, Y.; Nakano, H.; Urakawa, L.S.; Toyoda, T.; Sakamoto, K.; Yamanaka, G.; Sugimoto, S. Cold- versus Warm-Season-Forced Variability of the Kuroshio and North Pacific Subtropical Mode Water. Sci Rep 2023, 13, 256, doi:10.1038/s41598-022-26879-4.
Li, Z.; Guo, J.; Song, J.; Bai, Z.; Fu, Y.; Cai, Y.; Wang, X. Distribution, Movement and Generation Mechanism of the Mesoscale Eddy around the Kuroshio in the East China Sea (in Chinese). Journal of Marine Sciences 2022, 40, 1–10, doi:10.3969j.issn.1001-909X.2022.04.001.
Mason, E.; Pascual, A.; McWilliams, J.C. A New Sea Surface Height–Based Code for Oceanic Mesoscale Eddy Tracking. J. Atmos. Oceanic Technol. 2014, 31, 1181–1188, doi:10.1175/JTECH-D-14-00019.1.
Pegliasco, C.; Delepoulle, A.; Mason, E.; Morrow, R.; Faugère, Y.; Dibarboure, G. META3.1exp: A New Global Mesoscale Eddy Trajectory Atlas Derived from Altimetry. Earth Syst. Sci. Data 2022, 14, 1087–1107, doi:10.5194/essd-14-1087-2022.
Qiu, B.; Chen, S. Eddy-Mean Flow Interaction in the Decadally Modulating Kuroshio Extension System. Deep Sea Res. Part II 2010, 57, 1098–1110, doi:10.1016/j.dsr2.2008.11.036.
Teague, W. J., M. J. Carron, and P. J. Hogan. A Comparison Between the Generalized Digital Environmental Model and Levitus climatologies, J. Geophys. Res. 1990, 95(C5), 7167–7183, doi:10.1029/JC095iC05p07167.
Yang, H.; Lohmann, G.; Krebs‐Kanzow, U.; Ionita, M.; Shi, X.; Sidorenko, D.; Gong, X.; Chen, X.; Gowan, E.J. Poleward Shift of the Major Ocean Gyres Detected in a Warming Climate. Geophysical Research Letters 2020, 47, e2019GL085868, doi:10.1029/2019GL085868.
Zhang, T.; Li, J.; Xie, L.; Zheng, S.; Zheng, H. Statistical Characteristics and Path Analysis of Mesoscale Eddy in the East China Sea (in Chinese). Journal of Marine Sciences 2020, 38, 77–86, doi:10.3969/j.issn.1001-909X.2020.01.009.
Zhou, G.; Li, Z.; Cheng, X. Intrinsic and Wind-Driven Decadal Variability of the Kuroshio Extension in Altimeter Observations. Front. Mar. Sci. 2021, 8, 766226, doi:10.3389/fmars.2021.766226.
Zhou, S.; Zhang, Y.; Li, H.; Liu, L.; Liao, E.; Xu, F. Shoreward Shift of Oceanic Mesoscale Activity over the Last Three Decades. Nat. Commun. 2025, 16, 10381, doi:10.1038/s41467-025-65359-x.
Reviewer 2 Report
Comments and Suggestions for AuthorsMy review is attached. I recommend major revision.
Comments for author File: 
Comments.pdf
Author Response
Point-to-Point Responses to the Reviewers' Comments
Response to Reviewer #2:
Review of the manuscript climate-4122169 “Long-term variation of Mesoscale Eddy
Activity around the Kuroshio in the East China Sea during 1993–2023” by Mengrong Ding, Yujie Han, Yong Jiang, Yongheng Yao and Zipeng Yu submitted to Climate (MDPI)
Summary: This is a well-written paper that can be published pending major revisions. There is no problem with the text’s English, including grammar, style, spelling and terminology. There are no typos. The figures are legible. The references are complete and formatted properly. Major issues are all related to the lack of physical insight. Also, many trends are statistically insignificant, which is admitted in the text.
Response: Thank you for your thoughtful review and constructive comments on our manuscript. We appreciate your positive feedback and the specific, helpful suggestions you have provided. Your guidance has allowed us to significantly improve the manuscript's structure, clarity, and overall scientific narrative. We have carefully considered each of your comments and have revised the manuscript accordingly. A point-by-point response is provided below.
Recommendation: Major revision
Major issues (numerous references to literature are omitted for brevity and to comply with the MDPI policy regarding citations in reviews; I would be happy to provide a bibliography of most important papers if requested by the editor or authors):
The main problem is the complete lack of insight into different physical mechanisms that create different types of ocean eddies. The eddy detection algorithm used in this study shares the same critical drawback that is common to other eddy detection algorithms used in numerous studies. Namely, the radically different physical mechanisms that form different types of eddies are completely ignored. Every little bump of sea surface height (SSH) is declared an eddy. To illustrate the critical fallacy of this study it is sufficient to consider just two radically different types of mesoscale eddies. The first type is rings that are formed by meanders that grow, become unstable, and eventually pinch off. The Kuroshio Current forms meanders that produce such rings. The Kuroshio rings can be cyclonic and anticyclonic. The cyclonic rings usually contain cold water from the north. Therefore, cyclonic rings are usually called cold rings. The anticyclonic rings usually contain warm water from the south. Therefore, such rings are called warm rings. The cold cyclonic ring south of the Kuroshio Current and warm anticyclonic rings north of the Kuroshio Current are the largest, deepest, and most energetic mesoscale eddies in the World Ocean. Inside such rings, isotherms/isohalines/isopycnals are displaced by up to 800 m relative to the average background depth. Anomalies of SSH inside such rings are typically 40-50 cm. Such rings were investigated in numerous studies. The authors should familiarize themselves with such studies. The eddy detection algorithms should be tuned to be able to detect such rings and differentiate between the rings and other types of eddies.
Most currents, especially strong jet currents such as the Kuroshio Current, are accompanied by numerous eddies that form at the periphery of such current owing to the current shear. Such eddies have been first observed and reported near the Gulf Stream in the 1960s-1970s. They have been called spin-off eddies. The first papers about the spinoff eddies were published by Lee who observed such eddies in the Gulf Stream area. Spinoff eddies are often observed within peculiar meanders of western boundary currents. Such meanders have been originally called shingles owing to their overlapping pattern (like wooden shingles used in roof construction). The spinoff eddies are much smaller and less energetic than rings. Numerous in situ observational papers documented spinoff eddies in the Kuroshio Current region.
Another type of mesoscale eddies that can manifest as SSH highs are intrathermocline eddies (ITE). The ITEs are often called submesoscale vortices (SMV), although many ITEs are too large to be called submesoscale; they are of mesoscale. For example, the ITEs that form in the Strait of Gibraltar are typically >100 km in diameter. The ITEs are often called lenses owing to their characteristic shape. A review of SMVs was published by McWilliams in 1985.
Lately, other physical types of mesoscale eddies were reported that were observed in the Northwest Pacific, including the Kuroshio Current region. Obviously, satellite SSH data alone are not sufficient to elucidate the structure and dynamics of ocean eddies. In situ 3D (or better 4D) data are needed to produce a meaningful climatology of ocean eddies.
Response: Thank you for your thoughtful review and constructive comments. We fully recognize the importance of distinguishing different types of ocean eddies from the perspective of physical formation mechanisms, and this approach has crucial guiding significance for in-depth analysis of the enhancement process of eddy kinetic energy (EKE) in the study area. Eddies dominantly formed by different physical mechanisms exhibit significant differences in their energy characteristics and spatiotemporal evolution rules; clarifying such differences can indeed more accurately elaborate on the distribution of EKE and the driving logic behind its changes.
The core research focus of this study is to explore the spatiotemporal variation characteristics and overall enhancement process of EKE in the study area. As an important manifestation of mesoscale ocean motions, coherent eddies are only one of the important carriers of EKE energy, not the entire composition of EKE. Therefore, the analysis of EKE in this study is not limited to the scope of coherent eddies, but focuses on the overall variation rules of eddy kinetic energy in the entire study sea area.
Meanwhile, we also agree with the reviewer's judgment that relying solely on remote sensing data of sea surface height (SSH) from satellite altimeters is indeed insufficient to fully reveal the three-dimensional structure and deep dynamic evolution mechanisms of different types of eddies, which is an inherent limitation of researching the dynamic processes of ocean eddies using only satellite surface remote sensing data. Limited by the current acquisition conditions of multi-source in situ observation data, the core objective of the research design, and existing analytical methods, we are temporarily unable to carry out refined dynamic classification of eddies combined with three-dimensional (3D) or even four-dimensional (4D) in situ observations, nor can we quantitatively distinguish the differences in specific contributions of eddies with different formation mechanisms to EKE.
In future research, we will take this important suggestion from the reviewer as the core expansion direction. We will attempt to integrate multi-source observation data, such as in situ temperature, salinity, and ocean currents, carry out eddy classification research based on formation mechanisms, and quantitatively analyze the differences in contributions of different types of eddies to EKE in the study area. This will further improve our understanding of the physical mechanisms underlying the EKE enhancement process in the study area and make the relevant research conclusions more scientific and systematic.
Minor edits:
Line 53: “Over the past few decades, the ocean temperature in the ECS has generally
exhibited a rapid warming trend [15,16],” – Comments: On the climate scale (50-100 years), the sea surface temperature (SST) in the ECS exhibited a warming trend, which was extremely rapid during the final quarter of the last century. During that time, the ECS was the world’s fastest warming Large Marine Ecosystem (LME) as illustrated by a map from Belkin (2009) below that shows color-coded warming rates of SST in all 64 LMEs:
Fig. 2 from Belkin (2009). Net SST change (°C) in Large Marine Ecosystems, 1982–2006. Rapid warming (red and pink) is observed around the North Atlantic Subarctic Gyre, in the European Seas, and in the East Asian Seas, especially in the East China Sea. However, this climatic trend was punctuated by sharp reversals (also called regime shifts) as illustrated by a plot from Lee et al. (2021) that shows long-term variability of SST in the Taiwan Strait (southern ECS):
Fig. 4 from Lee MA et al. (2021). Long-term annual variability of the SST in the Taiwan Strait.
Response: Thank you for providing the key references and data support. This revision will enhance the accuracy of the background section of the paper. Indeed, scientific research requires such rigor. We have revised the description in the manuscript as follows to be more rigorous.
"Over the past few decades, the ocean temperature in the ECS has shown a marked warming trend, interrupted by sharp regime shifts, as reflected in the interannual and decadal variabilities previously documented for the Taiwan Strait. "
We have also cited the two papers you noted in the revised manuscript. Please refer to the revised version of the manuscript.
References:
Belkin, I.M. Rapid Warming of Large Marine Ecosystems. Progress in Oceanography 2009, 81, 207–213, doi:10.1016/j.pocean.2009.04.011.
Hirata, H.; Nishikawa, H.; Usui, N.; Miyama, T.; Sugimoto, S.; Kusaka, A.; Seto, T. The Kuroshio Large Meander and Its Various Impacts: A Review. J Oceanogr 2025, 81, 165–185, doi:10.1007/s10872-025-00753-z.
Round 2
Reviewer 1 Report
Comments and Suggestions for Authorsno more comments
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for addressing my concerns. I recommend acceptance in present form.
