Summary

The authors use a bevy of remotely sensed, reanalysis, and modelled products to determine the interplay between SST, Qnet, and MLD in the Mediterranean Sea from 1993-2022. More specifically, the authors investigate what the physical drivers are of MHWs during the onset and decline portions of the these events by decomposing the contribution of specific components of the mixed-layer heat budget equation o the anomalous SST observed during a MHW. These results are also parsed out by season, region, length of the event, and category (e.g. Moderate, Strong, etc.). The methodology used here is previously published, but has not yet been applied to the Med. The results presented here are an important contribution to the MHW literature and I think this paper is going to be very well cited over the coming years.

I have no substantive comments on the intro, methods (which is rare), or conclusions. I only have two minor issues (repeated in the specific comments below). The first is that the authors appear to not have accounted for the loss of shortwave radiation out of the bottom of the MLD in their mixed-layer heat budget equation. This will likely have little impact on the results, but it should either be accounted for, or the authors should argue for why they aren’t doing it. The second is that while I found the authors criticism of the methodology for how to account for the onset/decline periods, they unfortunately stop short of providing any sort of advice on what could be done instead.

All the best,

-anonymous

Title

- Consider “heat flux” rather than “heat fluxes”. Heat flux is generally used as a non-count plural noun. As the authors prefer.

Abstract

- ln 9: “research studies” is a tautology. Rather choose one word or the other.

- ln 10: “Marine Heatwaves” -> “Marine heatwaves” Or as the editor prefers.

- ln 16: “Moreover...” Which direction is the relationship? Is their more or less input of heat flux with increased MHW severity?

- ln 21: “have key” -> “have a key”

- ln 23: “emphasizes the need” Almost all studies of surface MHWs end by saying there is a need for the consideration of subsurface information. The criticism of the limitations caused by the onset/decline phase definition is interesting.

Introduction

- ln 49: “Darmaraki et al. (2023)” Hmm... I’m not sure a talk at a conference can be referenced here. This is not peer reviewed work. I don’t doubt the accuracy of the work though. The authors may want to edit the sentence slightly to account for the non-peer reviewed nature of these findings.

Data and Methods

- ln 58: It would be nice to list the horizontal resolution of the other products in this paragraph as well.

- ln 61: “ref. no. 05” Why not list the products in the text and the table in the same order?

- ln 85: Did you account for thew amount of shortwave radiation that passes out of the upper mixed layer?

Results

- ln 121: Reporting on MHW frequency is a bit of a tricky thing. This is because, particularly in the Med, we are beginning to approach the time when we have fewer MHWs as they start to last long enough to merge into one long MHW per year. I therefore always recommend to report on total MHW days per year, rather than frequency.

- ln 136: This sentence is confusing. After reading it a couple of times I understand what the authors are trying to say. It is unfortunately the nature of trying to explain this concept using words. It’s not easy to do and I don’t have a better recommendation.

- ln 249: “definition...” I agree with the authors to not give the full category definition here.

- ln 243: “events of higher severity categories...” I’m not convinced this is factually accurate.

Conclusions

- ln 248: “This...” Delete or merge this sentence with the next one. Not necessary to make this statement.

- ln 265: “These findings...” This is interesting, and makes me think of the basin-wide 2022 event. If oceanic advection is one of the primary drivers of the decline of events, but the whole basin is anomalously warm, and no wind exists to cause vertical mixing, then that would explain how a basin scale event can persist for months. That could be important for a future Med that will be much warmer than now.

- ln 727: “These...” I agree. But what do we do instead? It would be nice if the researchers could propose one or two ideas based on their experiences.

Table 1

I think it would be useful to give columns for the horizontal resolution, time step, and time series start to end date.

Figure 1

Change labels for “[deg]” to “[degC]” or somehow indicate it is degrees Celsius. Same for “[deg/year]”. If the longitude labels are only used for the bottom panels, then the latitude labels only need to be used for the left-hand panels. As the authors prefer. Why is a Mann Kendall test used and not a simple linear regression? Are the data significantly non-normally distributed? Generally speaking it is no longer preferable to use the rainbow colour palette. Rather used one of the viridis colour palettes instead (e.g. https://www.mathworks.com/matlabcentral/fileexchange/51986-perceptually-uniform-colormaps).

I like how the Atlantic coastline provides a natural little border for the panel labels :)

Figure 2

Panel a, colours should be switched for onset/decline to match the rest of the panels. Panel b, I think the comparison of onset vs decline will be communicated better if the bar plots are next to each other, rather than being stacked. Panels d-i, consider filling in, rather than line colouring, for the boxplots. It will make them look more substantial. I agree with the authors choice to limit the y-axis range for the boxplots to +-2 values, allowing longer tales to stretch outside of the plotting range.

Figure 3

Same consideration for boxplots as figure 2. Otherwise I like the layout of the panels. I am wondering if there are any spatial patterns across the Med that may explain some of the onset during MLD deepening MHWs, and vice versa?

Figure 4

Swap colours for onset and decline to match other figures. This is an interesting way of visualising these complex results.

Publisher’s note: the supplement to this comment was edited on 27 February 2024. The adjustments were minor without effect on the scientific meaning.