This is a nice analysis and a well-written paper, and will be of interest to ACP readers. I have only a few minor comments and suggestions.

Minor comments:

• Choice of 8 ppmv as a threshold for extrema. I know this is similar to what other studies have done (e.g., Schwarz et al) and that the authors results are not sensitive to the exact threshold per their sensitivity analysis, but would there be some value to using a threshold that varies regionally and/or with vertical level (e.g., a threshold of 3 sigma above the mean)? This might lead to better representation of “extreme” values and would also inherently take into account (im)precision of the MLS measurements. Perhaps some additional discussion/motivation or sensitivity analysis on this would be helpful.
• Line 41: “assent” -> “ascent”
• Lines 47-48: Consider moving parenthetical statement to end of clause, i.e. “… relative frequently over land (…), … “
• Line 58: “studying global” -> “studying the global”
• Line 67: I suggest using LMS here instead of LS to emphasize that LMS is the specific sub-region that is underestimated.
• Line 90-91: I think you are referring specifically to the MLS WV measurements here, not just MLS measurements in general, so you should probably mention that.
• Line 92: I recommend to use “less” here rather than “below” because below is often interpreted as a direction and could be cause confusion.
• Regarding the LMS SC extrema in Fig. 1. I was really confused by this when first mentioned on Line 157 because by your definition the LMS shouldn’t exist in the tropics. But you eventually provide an explanation around line 175. Maybe it’d be worth mentioning in section 2.3 that under some rare circumstances the criteria for LMS (340K < theta < 380 K, PV > 6, and 1 km above LRT) can be met in the tropics?
• Figure 4: Could some of these trajectories in the NP have come from NA or GC convection 10+ days earlier? There seems like a hint that the air came from all the way around the world at 10 days but perhaps the trajectories are too untrustworthy over that length of time.

The study presented by Tinney and Homeyer is scientifically interesting and significant to the community. Additional details and discussion regarding the methodology, however, would greatly strengthen the conclusions reached by the authors and improve the presentation. Please refer to the attached document for a detailed commentary.