Review of “Cold air outbreaks drive near-surface baroclinicity variability”

The authors present an analysis of the role that cold air outbreaks play in defining the baroclinicity over oceanic mid-latitudes (the North Atlantic and North Pacific basin). They analyse winters from a 40-year period using ERA5 reanalyses. They employ the relatively new and still novel isentropic slope framework to quantify the contribution to baroclinicity, finding that CAO play a substantial role in the Gulf Stream and Kuroshio regions. The key metric that supports this finding is the radio of CAO variance (normalised by the total variance) for the DIAB and TILT diagnostics, which can reach 40-45% for the Gulf Steam, even though the area of CAO is half that amount.

Overall, this is a concise, well written and nicely illustrated article. I don’t have any major points to address. However, I would say the article asks quite a lot of its readers. Those not so familiar with the isentropic slope framework don’t get a lot of help with interpretation here. This study builds on several previous ones outlining this framework and I wonder if a little more explanation would be beneficial here.

Specific Comments

1. I think the title needs a geographic qualifier. The authors have only examined two specific regions and I’d suggest that is reflected in the title. For example, “Cold air outbreaks drive near-surface baroclinicity variability over oceanic mid-latitudes” or “…over western boundary currents”.
2. Figure 1 – Panel (b) is a smorgasbord of greys. The continents, sea-ice areas and CAO masks are all different shades of grey! I’d recommend changing the CAO masks to the same shade of blue as the 4 K shading in panel (a).
3. Figure 1 – would it be useful to include the TILT term as panel (d) in this figure? It might give opportunity to explain these terms in a bit more detail for readers less familiar with this framework.
4. The ADV advection component of the slope tendency equation is neglected in this study. The reasoning behind this needs more explanation. The one given is “the interpretation is less straight-forward”. But I don’t think that is very satisfactory. Perhaps you can take recourse to previous work where this ADV term is shown to be less important in interpretation? What are the ramifications for neglecting this term?
5. I’d note that equations (7) and (8) could easily be merged.
6. I wonder if you could add a sentence or two more explanation for Figure 3 which shows the phase portraits. I wasn’t quite sure how to interpret these and also wasn’t clear how the slope shaded contours were plotted, i.e., how is the area of where they are plotted determined?
7. Figure 4 – I think the caption is incorrect here – the plots seem to have negative anomalies as thick contours and positive anomalies as thin contours? The caption talks about dashed contours.