the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Sep 2025
Marine Heatwaves across the central South Pacific: characteristics, mechanisms, and modulation by the El Niño Southern Oscillation
Abstract. Marine heatwaves (MHWs) are intensifying with climate change, endangering ecosystems such as coral reefs. Yet their regional characteristics and drivers remain poorly understood in many parts of the Pacific. Here we provide a comprehensive assessment of MHWs in the central South Pacific and across the five archipelagos of French Polynesia (representing ~77 atolls, more than half of world’s atolls, and more than 5 million km2 of maritime area, a region as vast as Europe), using sea surface temperature observations and an ocean reanalysis to investigate underlying mechanisms. MHW exposure varies widely across the region: its northern and southern parts (the Marquesas and Austral archipelagos respectively) experience the highest number of MHW days and the strongest cumulative intensities, especially during the warm season (Nov–Apr). In contrast, its central part (the Society, Tuamotu, and Gambier Islands) exhibits more moderate MHW characteristics. Heat budget analyses highlight the seasonally and regionally diverse mechanisms shaping MHWs. In central FP during the warm season, most of MHWs are driven by air–sea heat fluxes, while in the northern part, those driven by oceanic horizontal advection dominate. During the cold season, more MHWs driven by horizontal advection are observed in the whole region since the thicker seasonal mixed layer reduce the proportion of MHWs driven by air-sea fluxes. El Niño– Southern Oscillation (ENSO) strongly modulates MHWs occurrence: El Niño favors MHWs occurrence in northeastern FP, while La Niña increases MHW occurrence in the southwest with different spatial extent across ENSO flavors (Central and Eastern Pacific ENSO events). This modulation arises from reduced wind-evaporation cooling with reduced wind speed, shoaled mixed layers, and enhanced horizontal heat advection. These results greatly improve our understanding of MHW characteristics, dynamics and variability in this ecologically-fragile region.
- Preprint
(21693 KB) - Metadata XML
-
Supplement
(26502 KB) - BibTeX
- EndNote
Status: open (until 11 Dec 2025)
- RC1: 'Comment on egusphere-2025-4166', Anonymous Referee #1, 28 Oct 2025 reply
-
RC2: 'Comment on egusphere-2025-4166', Anonymous Referee #2, 12 Nov 2025
reply
Review of egusphere-2025-4166: “Marine Heatwaves across the central South Pacific: characteristics, mechanisms, and modulation by the El Niño Southern Oscillation”
General comments
This study provides a comprehensive assessment of the characteristics and drivers of marine heatwaves (MHWs) in the central South Pacific, in particular across the archipelagos of French Polynesia (FP). The authors use sea surface temperature observations for the period 1981-2024 to assess MHW metrics (frequency, duration, intensity) over FP, as well as their relations to ENSO and various ENSO flavors. An ocean reanalysis product for the period 1993-2024 is used to classify MHWs into types according to their main physical driver (e.g., air-sea heat flux or horizontal advection), and to investigate the relationship of mixed layer depth, wind stress, and sea surface height to ENSO and MHW occurrence.
Overall, this study provides a very thorough and comprehensive analysis of MHWs in the central South Pacific, which will be useful to regional stakeholders as well as to the broader research community working on MHWs at regional scales. The results are novel, the methods are clearly defined, and the presentation is easy to follow.
I recommend this manuscript for publication after the minor comments below have been addressed.
Specific comments
Ln. 23: “FP”: define abbreviation on first use
Ln. 29: “This modulation arises from...”: during which ENSO phase (El Niño or La Niña)?
Ln. 36: include another reference for future MHW projections (Oliver et al. 2018 is about MHWs in the past)
Ln. 33-56: the terms frequency, occurrence, and likelihood seem to be used interchangeably for the same concept (e.g., Ln. 35, 43, 45, 47). It could be useful to harmonize the terminology in order to clarify that the same thing is meant in each context
Ln. 55: “This highlights the need for regional studies...”: it also motivates the need for observation-based studies, as in this manuscript
Ln. 80: “Data and model simulations”: since no model simulations were performed for this study, the section could be called “Observational and reanalysis data” or similar
Ln. 128-133: It would be good to motivate the choice of this particular method to quantify cumulative heat stress
Ln. 154: “N phases”: define this (neural phases)
Ln. 195: “separately as well as their sum”: is this criterion also used for Q-MHW and HADV-MHW?
Ln. 194-197: does this classification assign every MHW to one of the categories, or does it leave out any events?
Ln. 202: “all MHWs [...] were used”: is this not a contradiction with the previous sentence?
Ln. 213: why are there “fewer MHW days” in the cold season compared to the warm season, is this because of the minimum duration of 5 days?
Ln. 241: “In the Austral islands, MHWs happen more regularly each year, with some bigger peaks”: this is imprecise wording, it is not clear what is meant by “regularly” (more even gaps? More frequent? What is the quantitative meaning of “some bigger peaks”?)
General: I recommend to avoid using the jet/rainbow colormap (e.g. in Figs. 1, 4, 6, etc.), see e.g. https://doi.org/10.5194/hess-25-4549-2021
Ln. 247: “the gray dashed areas”: these areas are not clearly visible. Maybe if a different colormap is used (as suggested above), the areas will become more visible. If not, perhaps an extra figure showing only the general study area as well as the archipelagos and their names could be useful (e.g. as a new Fig. 1).
Fig. 2: In the legend, spell out “warm season” and “cold season” (“warm” and “cold” could be taken to mean warm/cold extremes at first glance). In panel b, why are the mean values not given as in the other panels? In all panels, the mean values are not clearly legible. The colors for Gambier and Austral are hard to distinguish
Ln. 256: “The slope of the gray dashed line...”: it would be useful to test the statistical significance of these trends
Fig. 3: what do the errorbars in panel d) quantify?
Ln. 590: “Our analysis gives confidence in GLORYS, to study [the] vertical extent of MHW[s].”: how does this follow from the analysis?
Ln. 599: “TCs”: define abbreviation
Technical corrections
Ln. 17: “of world’s” -> of the world’s
Ln. 23: “most of MHWs” -> most MHWs
Ln. 25: “reduce” -> “reduces”
Ln. 33: “Marine heatwaves (MHWs), known as...”: it should be the other way around (XYZ events are known as MHWs)
Ln. 40: “...driving the MHWs” -> specify “driving MHWs over FP”
Ln. 42: “are known to be important drivers” -> is known to be an important driver (referring to ENSO)
Ln. 45: remove stray parenthesis “(“
Ln. 47: “intensity occurrence” -> intensity and occurrence? (or: intensity and frequency)
Ln. 54: “due to the resolution or not of...”: improve phrasing
Ln. 59: “due to FP geographical position”: due to FP’s ...
Ln. 60: “ENSO SST anomalies edges”: this is unclear
Ln. 60: “its effect”: specify the effect of what (ENSO?)
Ln. 107: “were computed” -> was computed
Ln. 111-113: the “number of events” and “gaps between events” are not really metrics for “each MHW”
Ln. 112: “SST change from start to peak”: is this the rate of change (in K/day) or really just the change (which would be equivalent to the intensity)?
Ln. 113: “MHW intensity can be expressed either...”: so which option did you choose?
Ln. 118: “In the various sections”: sections of the manuscript, or geographical/time sections? If the former, it may not be necessary to specify this
Ln. 124: “cooler-than-average SST climatology” -> SST cooler than the climatology
Ln. 136: It would be preferable to add an entry in the reference list for the cited website (including date of last access, etc.) and then citing it here. Also, mention what this data source is (“based on the French government’s XYZ”?)
Ln. 138: “Whether or not...”: the wording should be changed as this structure makes the sentence difficult to parse. E.g., you can simply start the sentence with “Removing a long term temperature trend...”
Ln. 175: “Qnet”: check formatting
Ln. 191: move the first minus sign on the RHS outside of the integral, specify which terms depend on the integration variable k, and maybe use a different symbol for the integration variable (e.g. t’ instead of k, as it represents time)
Ln. 215: “median values” -> median maximum intensity values
Ln. 220: “are found” -> is found
Ln. 238: “MHWs occurrence vary” -> MHW occurrence varies
Ln. 250: “see map on the right”: there is no map
Ln. 265: “1M km^2” -> 10^6 km^2
Ln. 267: Equivalent analyses of that...” -> equivalent analyses to those
Ln. 272: “that region”: what region?
Ln. 279: “EN, LN or N conditions”: it could be useful to remind the reader of the meaning of these abbreviations
Ln. 280: “using Pagli...” -> using the Pagli...
Ln. 335: “MMM”: what does this stand for?
Ln. 365: close the parenthesis
Ln. 366: “year”: remove
Ln. 380: “in surface”: at the surface
Ln. 388: “firstly”: mainly?
Ln. 398: “flux-driven”: does this refer to air-sea heat flux?
Ln. 425: “averaged across”: there is a word missing here
Ln. 439: “reinforces” -> reinforce
Ln. 464: “smooth out eddies pattern” -> smooths out eddy patterns
Ln. 473: “present” -> presents
Ln. 474: “hardly interpretable” -> hard to interpret?
Ln. 498: “back to normal” -> neutral/zero
Ln. 509: “to represent them”: remove
Ln. 523: “and may also reduce” -> and potentially also reducing
Ln. 572: “the most extreme events”: globally?
Ln. 600: “intra-seasonal, variability”: remove comma
Ln. 626: “relating the” -> relating to the
Ln. 632: “While OISST dataset” -> while the OISST dataset
Ln. 633: “some MHWs commonly used metrics” -> some commonly used MHW metrics
Fig. S1: the colorbar label shold be degrees C, not “thresh”
Citation: https://doi.org/10.5194/egusphere-2025-4166-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,119 | 88 | 16 | 1,223 | 59 | 37 | 31 |
- HTML: 1,119
- PDF: 88
- XML: 16
- Total: 1,223
- Supplement: 59
- BibTeX: 37
- EndNote: 31
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of OS manuscript 2025-4166 submitted to Ocean Science
The manuscript entitled “Marine heatwaves across the central South Pacific: characteristics, mechanisms, and modulation by the El Niño Southern Oscillation” by Bastien Pagli et al., submitted to Ocean Science, investigates the contribution and importance of El Niño Southern Oscillation (ENSO) to marine heatwave occurrences and evolution in the central South Pacific and across French Polynesia.
Overall, I find the manuscript to be a valuable contribution to our understanding of ENSO on marine heatwaves across this region that is sensitive to ENSO phase both temporally and spatially. I recommend the manuscript be potentially acceptable for publication once the comments below have been satisfactorily addressed by the authors.
Comments and Concerns
L2: In the title, I recommend deleting “the” as with “the El” it’s like a double-“the”
LL18-19: “MHW exposure varies widely across the region” – at this point, the reader is unaware what the authors mean by “exposure” here, which is a term most often used when referring to species and/or ecosystem exposure to a hazard
L23: “FP” acronym is undefined; delete “of”
L24: Is there really a “cold season” across French Polynesia?
L26: change “MHWs occurrence” to “MHW occurrences”, here and throughout the manuscript
L33: “known as prolonged periods of extreme ocean temperature” – MHWs are not periods of extreme temperature (which would be measured in time units), but rather they are temperature extremes that persist (measured in temperature units)
LL35-36: “and are projected to continue rising in the future (Oliver et al., 2018)” – this last part of the sentence is not covered by the cited reference which focuses on historical data, and not projections. The authors should cite another reference for this.
L36: “Developing skillful … their impacts.” – This is a leap in a single sentence. It needs to be explained further. Skillful forecasts do not necessarily translate to reduced ecological impacts.
LL37-38: “Such forecasts … variability.” – A relevant paper to cite here is Holbrook et al. (2020, NREE) which makes and builds the case for understanding MHW predictability and prediction.
L43: “increased MHW” – also include “and suppressed MHW”
L45: “(Gregory et al., (2024)” should be “Gregory et al. (2024)”
L47: acronym “SST” is undefined
L53: “MHWs” should be “MHWs’ “
L59 and L64 and throughout: I consider it more appropriate to replace “impacts” with “influences”, since the word “impacts” is more appropriate for socio-ecological impacts, while “influences” is between the physical components
L81: “OISST” – write in words first and then use the acronym
L82: delete “resolution” – the data are interpolated to a 0.25o grid, which does not necessarily reflect the true “resolution” of the original sampling
LL88-89: “Due to the known … ERA-Interim” – why not use ERA5 throughout?
LL138-140: “Whether or not … Amaya et al., 2023; Capotondi et al., 2024).” – This is not a sentence, please rewrite. Also, note that the Amaya et al. (2023) paper was challenged by Sen Gupta et al. (2023, Nature) and which helped lead to the paper by Smith et al. (2025, Prog Oceanogr). It would be good to reflect this in the manuscript.
LL140-141: “Both approaches are complementary … address.” – The key reference here is Smith et al. (2025, Prog Oceanogr).
LL141-144: Another approach would be to follow the guidance of Smith et al. (2025), as the terminology “total heat exposure” suggests that species are indeed exposed to the hazard. I’m not sure we can absolutely assume that?
L179: acronym “MLD” is undefined
LL184-186: the word “net” should be explicitly used in each term in the equation, e.g. “LHF the net latent heat flux” etc.
L191: In Equation (5), why is “dk” used rather than “dt”?
L199: “sea-level (SSH)” – should this be “sea surface height”? Sea level and sea surface height can be subtly different.
LL210-212: Following the Hobday et al. (2016) definition, why isn’t the average total number of MHW days/year close to 10% of the year (~36/year) everywhere?
L213: “During the cold season … compared to the warm season.” – Based on the 90th percentile seasonally varying threshold applied using Hobday et al. (2016), why are there less cold season MHWs than in the warm season?
L233: “cumulative impacts” should be “cumulative intensities”
L250: Figure 2 caption – “map”, I think should be “legend”
LL273-274: “However some differences … than in OISST.” – This is consistent with the analysis of Pilo et al. (2019) based on ACCESS-OM2 simulations across different model resolutions. I suggest this would be an appropriate reference to cite as context.
L394: Figure 6 choice of colorbar for panels a-c and e-g make it difficult to discern proportions <40%.
L542: Conclusions section usually follows the Discussion section.
L568: I think “types” would be better used to replace the terminology of “categories” in the context used here, as “categories” is typically reserved for measures of MHW intensity.
LL605-606: Another earlier study of MHWs in the region and relevant reference here is Holbrook et al. (2022, Glob Planet Change).
L632: “Gupta and Sil (2024)” should be “(Gupta and Sil, 2024)”
References
Holbrook NJ et al., 2022: Impacts of marine heatwaves on tropical western and central Pacific Island nations and their communities. Global and Planetary Change, 208, 103680, https://doi.org/10.1016/j.gloplacha.2021.103680.
Holbrook NJ et al., 2020: Keeping pace with marine heatwaves. Nature Reviews Earth and Environment, 1, 482-493, https://doi.org/10.1038/s43017-020-0068-4.
Sen Gupta A et al., 2023: Marine heatwaves: definition duel heats up. Nature, 617, 465, https://www.nature.com/articles/d41586-023-01619-4.
Smith KE et al., 2025: Baseline matters: Challenges and implications of different marine heatwave baselines. Progress in Oceanography, 231, 103404, https://doi.org/10.1016/j.pocean.2024.103404.
Pilo G S et al., 2019: Sensitivity of marine heatwave metrics to ocean model resolution. Geophysical Research Letters, 46, 14604-14612, doi:10.1029/2019GL084928.