the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
On the drivers of Mediterranean Marine Cold Spells under Climate Change
Abstract. Marine Cold Spells (MCSs) are prolonged, extreme cold temperature events in the ocean that can disturb marine ecosystems. Here, we analyze for the first time historical (1980–2014) and future (2015–2100) MCSs in the Mediterranean Sea, using the output from a high-resolution regional climate system model, under global warming levels 1–4°C and the 25 high-emission climate change scenario SSP5-8.5. Historically, atmosphere forcing dominates open-ocean MCSs in all seasons, except winter, where Advection plays a more dominant role across the basin, reflecting the influence of alongshore currents and mesoscale variability in dynamically active regions. Mixing drives coastal and localized events mostly in the summer, likely due to regional wind forcing. As global temperatures rise, projections relative to a fixed historical baseline indicate up to 92 % decline in future MCS occurrence, up to a 50 % reduction in their duration and up to 19 % decrease in 30 their intensity already by 2036–2055. These changes are more pronounced during winter, with the events becoming increasingly confined to the northern Mediterranean and progressively disappearing. When defined relative to a shifted climatology, future MCSs represent extreme cold events of a warmer climate and are projected to become less frequent, slightly longer and more intense compared to the past by the end of the century. The driving mechanisms of both types of future MCSs shift from a dominant atmospheric control in the historical period towards an increasing influence of oceanic 35 advection and mixing, particularly in the summer, autumn and spring and by the end of the century.
- Preprint
(2508 KB) - Metadata XML
-
Supplement
(4052 KB) - BibTeX
- EndNote
Status: open (until 10 Jun 2026)
- RC1: 'Comment on egusphere-2026-962', Anonymous Referee #1, 15 May 2026 reply
-
RC2: 'Comment on egusphere-2026-962', Anonymous Referee #2, 25 May 2026
reply
General Comments and Summary
This manuscript provides a valuable and necessary investigation into the physical drivers of Mediterranean Marine Cold Spells (MCSs) under historical and future climate scenarios. Using a mixed-layer heat budget (MLHB) to decompose forcing mechanisms (atmosphere, advection, and mixing) offers insights into how extreme cold anomalies evolve in a warming ocean. The dual-baseline approach effectively demonstrates the shifting nature of future extremes. The study is concise and well-written. I recommend the manuscript for publication after a few minor clarifications and visual refinements are addressed.
Specific Comments
- In Section 3.1, the spatial consistency between modeled and observed surface MCS characteristics is noted as weak, attributed primarily to differences between the 0.5 m near-surface model temperature and satellite skin-layer measurements. Since grid interpolation alone is unlikely to cause a discrepancy of this magnitude, it would strengthen the paper to briefly expand on this.
- The manuscript utilizes historical observational satellite data ending in 2017. The "Mediterranean Sea - High Resolution L4 Sea Surface Temperature Reprocessed" product is available up through 2026. Given that basin temperatures have been on a steep, documented rise in recent years, it would be interesting to discuss (at the authors' discretion) whether the projected trends—specifically the initial reduction or regional disappearance of MCSs—are already beginning to emerge in the actual observational record over the last decade.
Technical Corrections
- General Figure Comment: For the multi-panel figures, the darker gray tone used for landmasses competes visually with the sea, often requiring the reader to zoom in. Using a lighter gray shading for the land would make it much easier to focus on the marine data.
- Figures 1 & 2: Ensure panel titles are consistent in capitalization (e.g., "WINTER").
- Figure 5: Using white as a value color makes it difficult to distinguish between areas with low event counts and regions where MCSs are completely absent/masked out.
- Figures 6 and 8: The visual dominance of red across most panels is confusing for a relative fractional shift metric. It visually implies a near-uniform increase across all drivers simultaneously. If regional weightings or absolute frequencies cause this apparent contradiction, it would help to clarify this in the text/caption, or to display the absolute base frequencies alongside the changes. Alternatively, refining the color shading (better contrast perhaps) in Figures 5 and 7, where the base frequencies are shown, could help resolve this visual ambiguity.
- Table S6 (Supplement): The column header "GWL3" is visually misaligned.
- Line 44: Capitalize the "W" in western Mediterranean (WMED).
- Line 144 (Equation 1): The MLHB equation contains a typographical artifact in the forcing component denominator "𝝆𝟎𝑪𝒑𝒉". Please verify the typesetting.
Citation: https://doi.org/10.5194/egusphere-2026-962-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 766 | 359 | 75 | 1,200 | 224 | 68 | 134 |
- HTML: 766
- PDF: 359
- XML: 75
- Total: 1,200
- Supplement: 224
- BibTeX: 68
- EndNote: 134
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Summary
The authors start with a thorough and interesting review of known MCSs that have occurred in the Med, what their physical drivers were, and if they had any bio/ecological impacts. They then show the validation of their model against satellite observations. With an issue being that the authors use MLT from the models, but SST from satellite obs. This mismatch is acknowledged, but not much more is said about it before going into the results on the future MCS state in the Med. Using both fixed and moving baselines, the authors show what the current/historic forcing of MCSs in the mixed layer have been, and what they may become. Interestingly, regardless of whether the baseline is fixed or not, the changing to the drivers of MCSs will be similar moving into the future. That being that extreme cold will come primarily from advection or mixing, and not from direct surface heat fluxes. Using a fixed baseline, the authors also provide maps of where MCSs will first begin to disappear from the Med.
The study is well written, concise, and provides very interesting results. The results are rather dense throughout, making it a bit difficult to keep everything straight in ones mind, but that is difficult to avoid for a study of this nature. A lot of information is provided, and many interested readers will be able to extract a lot of relevant information from this study. The authors may want to consider putting their results up on a public repository somewhere so that the data may be downloaded and used for future studies y other teams.
I have no substantive comments and therefore recommend this article for publication more-or-less as is.
Title
- Why ‘On the drivers of...’? Why not just start with ‘The drivers of…’?
Abstract
- Very interesting. Very succinct.
1. Introduction
- ln 60: ‘MSC’ → ‘MCS’
2. Methods
- It’s an interesting idea to analyse MLT rather than SST.
– Though as one may see in the model vs satellite results, this likely has a large impact on the comparability of these two data sources. The authors acknowledge this but don’t provide a solution. Ultimately though I don’t think this is an issue as the analysis of MLT is a good idea. Especially for a mixed layer heat budget analysis focussed study.
- Also interesting to use the GWLs as the reference for the shifting baseline.
3. Results and Discussion
- ln 214: “A similar…” I don’t think the interpolation of satellite data is causing the mismatch. Rather the fact that MLT and SST are being compared.
- ln 351: “To our knowledge…” I agree.
- ln 434: “In caution” → “with caution”
4. Conclusion
- 573: “although spatial consistency with the observed MCS count and duration is weak” Likely because the authors are comparing SST with MLT.
Appendix
The last page appears to be blank.
Table 1-2
No comments.
Figure 1
A diverging colour palette isn’t really appropriate for linearly increasing set of values (i.e. 0 to 30). Though the current palette is easy to read. As the authors prefer.
Figure 3
It’s interesting that the count of MCSs between datasets over seasons are so different, while the intensities are so similar.
Figure 4
It’s interesting how clear the patterns are.
Figure 5
I recommend changing the colour palette for the first set of panels. Use something that doesn’t have white as one of the values so it is more obvious which regions have no MCSs.
Figure 6 & 8
I’m not really sure I understand what these figures are showing. So red means that the future will have a greater percent of the MCS affected by forcing, advection, or mixing? How can everything be red? If this is showing percentage, shouldn’t there be an equal amount of blue somewhere?
Figure S1-11
No comments
Table S1-S7
No comments.