Summary

This study investigates using a model hierarchy the effect of climate change onto persistent cold-air pool episodes (PCAP) in the Grenoble valley system. Such events are associated with strong vertical stability and thermal inversions inside the valley themselves, and with poor air quality close to the surface during winter. Thus, studying such events can be of interest to the population living in such valleys in Grenoble and around the world. Global warming can have two competing effects on the strength of thermal inversions: surface-based warming can reduce their strength, while mid-tropospheric warming can increase it. Local conditions can make one process prevail on the other, with opposite effects. The seamless interplay between global climate change and local weather discussed in the paper fits well the scope of Weather and Climate Dynamics.

I find the manuscript well written and easy to follow, even for people (as the reader) who are not so expert in valley cold pools. The motivation is clearly provided in the Introduction, the methodology is explained in great detail, the results are exposed in a concise manner and generally back up the conclusions. I have some minor comments about the interpretation of the results, and would recommend this manuscript for publication after they are satisfactorily addressed.

General comments

1. The importance of surface warming as a driver of reduced stability during inversions is pointed out several times (line 336, 449, 472) and is important for the interpretation of the results. However, the detailed reason for such an increase of temperature during PCAP episodes is not explained clearly in the manuscript. The reference of Bailey et al. 2011 is provided at line 337 but I was not able to find relevant explanations there from a quick read. On the other hand, Fig. 9 shows an increase of around 4K for both surface and top of inversion (lines 369-374), and this might be indicative of warmer air at all levels due to advection, without obvious changes in stability. The authors need to point out the processes that would (eventually) enhance such a surface-based warming: for instance, are higher temperatures at the surface maybe due to changes in cloud cover, that modify the short-wave (daytime) or long-wave radiation balance?
2. According to the performed simulations, the vertical stability seems to increase for stagnation events in a warmer climate, especially for the elevated thermal inversions (Fig. 11). This hints to the presence of warmer air above the inversion in the 2043 event, likely resulting from non-local processes such as advection. Together with a reduction in the height of the BL of ~100m, the model simulations would thus indicate a strengthening of thermal inversions, related mostly to the warming at upper-levels. This would be in contrast with the current interpretation of the results, which indicate an average reduction in inversion strength during the next century driven by surface-based warming. Can the authors please reconcile these contrasting results? Are there reasons to expect extremes to follow a different trend than the mean?
3. Given the fact that VHD does not appear to change substantially between the two events, can we conclude that what changes for PCAP events in a warmer climate is not the rate of cooling but rather the initial temperature at which cooling starts?
4. It is not yet clear how the projected trends in inversion height and strength will reflect themselves in air quality, and some apparently contradictory statements are found in the manuscript. For instance, at lines 474-476 is written “the less stable winter atmosphere could positively impact the air quality”, but previous results (summarized at line 460) indicate that future episodes will feature “a lower inversion height”, which would worsen air quality. The contradiction is mostly between the effects of inversion strength and inversion height, but I would think that the latter is more important than the former, provided that a sufficiently strong inversion does not “break” during daytime.

Technical/Typos/Etc…

Line 226: which value has been “rounded” and how?

Line 245-246: the definition based on 4 weather regimes confounds together Scandinavian blocking, European blocking and sometimes even Greenland blocking, that other authors would indicate as responsible of stagnation events over central Europe (e.g., 10.1088/1748-9326/ab38d3). Please provide reference(s), or indicate that this statement only refers to the k=4 choice for weather regimes.

Line 269: at which level is the wind estimated?

Line 381: “wind intrusion” is not commonly used and might be misleading, please use other formulations depending on the meteorological object associated with that wind maximum (e.g., is it a jet streak, maybe related to a small upper-level trough?).

Lines 425-427: this sentence is not clear, because if the assumption is “not correct”, how can the assumption “approximately hold”? Please specify why the linear assumption is not anymore a good one for Ep2043, and whether the linear extrapolation to compute VHD is still useful.

Line 447: scenario corresponds to a statistically

Line 468: in which sense “become not significant”? This terms should be used only in the context of statistical tests.

Table 2: add asterisks or denote otherwise which changes are significant

Fig. 4, 5, would profit of small titles indicating which quantity is being looked at (e.g., “PCAP duration”, “PCAP stability”, etc…).

Fig. 9, 11: mark the three days period chosen for averaging in both figures (lines 402-403).

Fig. 11: epThe

The study investigated persistent cold-air pools (PCAPs) in the Grenoble valley using statistical analysis with climate model data to drive a regional climate model. The authors also conducted a detailed comparison of two PCAPs, one from the recent past and one from the mid-21st century, using simulations from a high-resolution model. PCAPs adversely affect air quality, making them significant for society. Therefore, understanding their future changes holds general importance. The paper is well-written and presents comprehensive methods and reasoning. The climatology offers valuable insights into future PCAP changes.

However, a main concern is the limited sample size for comparison, with only two chosen episodes. While the authors provide reasonable justification, conclusions from this small dataset must be treated cautiously. The study highlights substantial variability between cases. The authors infer a smaller inversion height in the future. However, the difference in mean inversion heights between future and historical cases appears insignificant (as indicated by Table 3, i.e. fall within standard deviation ranges). Furthermore, the two PCAP episodes differ in nature (e.g., one is impacted by initial cold air, potentially influencing its development). Probably the only secure conclusion is that episodes will be generally warmer in a warmer climate. Determining future structural changes from only two simulations is not possible; a larger sample is imperative. Nevertheless, this study marks a crucial advance towards a broader examination of PCAPs using high-resolution models.

Minor points:

• Lines 23-38: The first paragraph of the introduction could be improved. It reads a bit bumpy. Moreover, I think that you should describe the implications of PCAPs on society in more detail and introduce the relevance of your study for society, i.e. why it is important to study this phenomenon.
• Line 36: "The qualificative ground-based,..." - can you please rewrite the sentence or put the terms ground-based, surface-base etc in italics? This would be easier to understand.
• Line 117: "(CMIP6, Eyring et al. (2016))" - should be: (CMIP6, Eyring et al., 2016)
• Line 189: Is "Noah Land surface model" correct? Moreover, I cannot find the reference Chen and Dudhia (2001) in your reference list!
• Figure 6: Can you try to plot a violin plot or a box plot. This could improve your plots. It is not easy to grab information from these pure scattered points.
• Figure 7: Can you please again explain in more detail: why do you see the significance of trends for larger periods, but not for the smaller ones?  
• Figure 8: Do your trends stay significant if you choose different periods, for example 2020-2080? The variability seems so high, that I wonder what will happen, if you change periods (for example by rolling over 20 years).
• Figure 9: It is really difficult to see the warm temperature values on the left figure. I understand, that your goal is to show that it is much warmer in the future. However, it would be great if you can additionally plot the figure by normalized values (for example between minimum (=0) and maximum (=1)? This figure could also be added to the supplementary material if you do not want to add another figure to your main paper.
• Line 391ff, equation (4) and line 425-427: Can you please explain why you do not expect much difference if you set z_t to 1500 m or to the inversion height (if I understood it correctly?). Maybe you can give some simple, idealized examples (for different vertical profiles) to explain your assumption?
• Line 409 (see also Table 3): You say that the inversion height is on average 100m lower in the future, however, this is not significant! Please clarify!
• Line 410: "Consistent with these observations ..." - Why is this congruent? In the preceding sentence, you noted a lower inversion height, while here, you mention a similar Vertical Heat Diffusivity (VHD). Can you please provide clarification?
• Lines 425-427: I do not understand the sentence starting with "Figure S8 [..]". Please explain.
• Line 460: I think that you cannot draw the conclusion that the inversion height in the future is generally lower. Please clarify that you just looked at two example periods. Since the variability between episodes is large, in my opinion it is not possible to draw such conclusions.
• Line 486: There is a word missing in the sentence.