Frontal Coupling Regulated by Mesoscale Eddies in the Subarctic Frontal Zone

Sun, Aoran; Zhang, Chunling; Yang, Huangxin; Zhou, Yuntao

doi:10.5194/egusphere-2026-4384

Preprints

https://doi.org/10.5194/egusphere-2026-4384

Preprints

04 Aug 2026

| 04 Aug 2026

Status: this preprint is open for discussion and under review for Ocean Science (OS).

Frontal Coupling Regulated by Mesoscale Eddies in the Subarctic Frontal Zone

Aoran Sun, Chunling Zhang, Huangxin Yang, and Yuntao Zhou

Abstract. Frontal coupling and its interaction with mesoscale eddies plays a critical role in water mass exchange and the ecodynamic balance in the Subarctic Frontal Zone of the northwestern Pacific. Using ten years (2014–2023) of daily satellite observations, this study systematically investigates the coupling characteristics of temperature and chlorophyll fronts under eddy modulation, along with the spatiotemporal responses of frontal coupling to eddy evolution by eddy partitioning, normalization, and lagged cross-correlation and mediation analyses. The eddy modulation on temperature and chlorophyll-a fronts pronouncedly depend on eddy polarity and developmental stage. Cyclonic eddies enhance chlorophyll-a fronts in the eddy core and concentrate temperature fronts at their periphery, whereas anticyclonic eddies suppress both frontal types in the core with recovery in the exterior. Under cyclonic eddy perturbation, the probability of frontal coupling increases with eddy intensity. The optimal lag time is 3–8 days, and the coupling probability in the eddy core is much higher than that of exterior. Anticyclonic eddies show a weaker promoting effect that is inversely related to eddy intensity, with a lag time of longer than 10 days. The responses of the two frontal types to eddy forcing are asynchronous. There is a time lag of several to over ten days between the physical-biological frontal coupling and eddy generation. While temperature fronts play a significant mediating role in this process, the mediation mechanisms differ according to eddy dynamic parameters and lag time. These findings demonstrate that eddy polarity and intensity jointly determine the direction, magnitude and duration of frontal coupling. That will provide a better understanding for the coupled mechanisms between physical and ecological fronts, as well as the oceanic dynamical and biological responses to eddy modulation.

Received: 21 Jul 2026 – Discussion started: 04 Aug 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Aoran Sun, Chunling Zhang, Huangxin Yang, and Yuntao Zhou

Status: open (extended)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-4384', Anonymous Referee #1, 27 Sep 2026 reply
The paper uses satellite data and reanalysis product to detect and characterize cyclonic (CE) and anticyclonic eddies (ACE) at the Subarctic Frontal Zone. Specifically, the authors aim to further understand how temperature and chlorophyll-a (Chl-a) fronts interact and how this interaction is modulated by CEs and ACEs. What seems new is the step from tracer values to tracer fronts: whether eddies affect where and when SST and Chl-a fronts coincide. I believe this is an interesting and relevant scientific question, and the ten-year daily dataset is a valuable basis for addressing it. Some results (e.g. Table 1) are consisten with known eddy composites and are encouraging. However, as currently written and presented, the methodology is hard to follow, and it is difficult to understand how the authors arrive at their interpretations. Key methodological descriptions are missing (listed below), which makes the figures and results hard to evaluate. Many definitions and decisions are left unexplained (see concerns below) in some cases a citation is provided without explaining how the cited method was applied in this study. Some figures do not appear to show what the text describes (specifically Figs. 6 and 7). In particular, the sign and structure of the relative vorticity of the eddies described in the paper are unusual and would need further explanation in comparison with typical eddy studies. Although I believe the scientific idea is a very good one, the manuscript needs substantially more methodological information, and some additional analysis, before it can be considered for publication.
Major concerns
Data set and Methods: The following issues made it difficult to interpret and assess the results:

The authors describe their data as "three categories of remote satellite observation" (line 66), but the surface velocity used for eddy detection comes from a reanalysis (GLORYS12V1; Lellouche et al., 2021, Line 355). A reanalysis is a model constrained by observations, so detected eddies may be displaced relative to features in the observed SST and Chl-a fields. Moreover, since GLORYS12V1 assimilates satellite SST, the eddy and SST-front variables are not fully independent. Please clarify this choice and why altimetry-derived velocities were not used, as in other eddy studies, including those applying the same Nencioli et al. (2010) detection method (e.g., Chelton et al., 2011; Gaube et al., 2014; Dong et al., 2022).

The datasets have different resolution (0.01° SST, 4 km Chl-a, 1/12° velocity), but no explanation is given of the common grid, how the interpolation was done, or why these choices were made. This is important to better understand how the gradients are compared between properties

How were eddies selected? Please provide the size and lifetime distributions, and state whether any minimum radius, amplitude, or lifetime threshold was used to discard eddies detected by the algorithm. The Nencioli et al. (2010) parameters should also be reported. Figure 1 shows many very small eddies spanning only a few grid cells.

The authors discuss the eddy life cycle, but there is no description of how eddies are tracked. In addition, if composites are built from daily eddy snapshots, how do the authors account for the autocorrelation of eddies observed on consecutive days? The effective sample size is likely much smaller. Also how did the authors arrive at an n = 712 (Table 1).

Eddy kinetic energy is not defined, including how the mean flow is defined. It is also unclear how the EKE levels used in Sect. 3.1 to classify eddies were defined, and how they relate to the percentile classes in Fig. 2.

For the Pearson correlation methodology, what is a "valid grid cell"?

How do theta count for front orientation besides coincidence of location

"Lag day" is not defined consistently across Figs. 4, 5 and 8, and no information is given on how many eddies "survive" to longer lags (e.g., day 16). At longer lags only long-lived eddies remain, so changes with lag may partly reflect a changing population rather than eddy evolution. This made the lag results hard to follow or review further.

Results and Discussion: The concerns about the methods made the results hard to review, beyond the points listed below. In general, the results would benefit from a more careful description of the steps taken to obtain the plotted values, and the assumptions made need to be justified.
Section 3.1 Figure 2 would benefit from showing the distributions of these paired differences rather than absolute values, since the background temperature range makes it hard to see differences between the inside and outside of the eddy. The anticyclones have a lower mean SST than the cyclones, which likely reflects their locations north and south of the Kuroshio Extension rather than an eddy effect. A seasonal and/or regional breakdown would help separate eddy effects from background variability.

Line 120: The authors mention "eddy mixing" without explaining what they mean. Mixing would reduce, rather than create, differences in environmental properties between core and exterior. Do the authors mean stirring or trapping?

Section 3.2: The data availability section indicates that the Chl-a product is Copernicus-GlobColour L4 daily, which is gap-filled. Temporal interpolation across cloudy days both creates autocorrelation over several days and can make a front "emerge" simply when new cloud-free data become available. I may be incorrect in this assumption, but since no information is given about this processing, it is hard to judge whether the lagged analyses (FGP and Fig. 5) reflect real frontal evolution. Distinguishing missing from interpolated data, or repeating key results with L3 data, would address this

Figure 3: shows the most interesting results. The authors should discuss how they compare with other studies, for example the peripheral stirring signal described by Chelton et al. (2011) and Gaube et al. (2014).

Section 3.3 is very hard to follow. The authors describe a "typical quadrupole structure" in the relative vorticity of both CEs and ACEs, which I have not seen in the literature for eddy vorticity composites. A citation and explanation are needed for why these eddies lack the typical positive core (CEs) and negative core (ACEs). Moreover, during the formation stage of cyclonic eddies the authors state that "negative vorticity dominates the eddy core" (line 229), whereas Northern Hemisphere cyclones have positive relative vorticity. An azimuthal quadrupole centred on zero, as in Fig. 6a, is characteristic of a strain component rather than vorticity. The composites are also nearly identical across life stages, despite the expected changes in eddy amplitude. Please define how relative vorticity was calculated (formula, sign convention, and the ordering of the zonal and meridional array dimensions). Because relative vorticity is also the predictor in mediation Model 2, all Model 2 results in Table 2 and the Appendix depend on this check

In general, the manuscript uses causal language without strong mechanistic evidence. The authors mostly provide correlations, not demonstrated effects. They should be more explicit about why they believe the proposed processes are the ones driving these correlations, and justify their assumptions. For example, "eddy regulation" of frontal coupling is not defined. The results show that FGP and lagged front-intensity correlations differ with eddy polarity and EKE, which establishes an association rather than regulation. The two candidate mechanisms introduced at lines 180–183 (strain-driven frontogenesis vs. eddy stirring) are not tested
Minor comments
The main question if the paper is not clearly stated
“However, It remains unclear that the regulation of eddies in the coupling relationship between temperature and Chl-a fronts, as well as their spatiotemporal covariation”. To investigate this question.. (line 55): The terms "regulation" and "coupling" are not defined,
Figure1:A geographic reference is needed; the reader is left to find a map to check exactly where the study area is.
The word coupling is used without being defined. What exactly is meant by coupling between temperature and Chl-a fronts: co-location, correlated intensity, or lagged formation? The term is used for both lagged front formation (FGP) and intensity co-variation (Fig. 5). In addition, "coupling" usually describes a two-way dynamical interaction (e.g., air–sea coupling), whereas chlorophyll fronts do not feed back on temperature fronts. A term such as "frontal co-location" or "frontal correspondence" may be more appropriate.
The manuscript contains numerous typographical errors and several incomplete sentences (e.g., lines 55–56, 135 and 336), which in places make the meaning difficult to follow. A careful proofreading of the full text is recommended.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4384-RC1

Aoran Sun, Chunling Zhang, Huangxin Yang, and Yuntao Zhou

Viewed

Total article views: 182 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
107	54	21	182	17	14

HTML: 107
PDF: 54
XML: 21
Total: 182
BibTeX: 17
EndNote: 14

Views and downloads (calculated since 04 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	42	18	7	67
Sep 2026	54	26	12	92
Oct 2026	11	10	2	23

Cumulative views and downloads (calculated since 04 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	42	18	7	67
Sep 2026	54	26	12	92
Oct 2026	11	10	2	23

Viewed (geographical distribution)

Total article views: 151 (including HTML, PDF, and XML) Thereof 151 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 06 Oct 2026

Short summary

Temperature and chlorophyll-a fronts in the northwestern Pacific are modulated by ocean eddies. Analysis of daily satellite data from 2014 to 2023 shows that eddy polarity and strength control frontal coupling intensity, while chlorophyll-a fronts respond several days later than temperature fronts. This time lag indicates that biological responses to eddy forcing are inherently slower than physical adjustments.


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