|I think the authors have responded reasonably to most of my comments, except perhaps one point which I don't quite agree with their response. Other than that, I only have a couple of minor comments.|
1) In my previous comments, I commented on the missing troughs and ridges along the subtropical jet south of the Tibetan Plateau, which are clearly present in previous studies (e.g. Fig. 14 of Hoskins and Hodges 2002, Chang and Yu 1999, and Chang 2005). In fact, if one compares the schematic in Hoskins and Hodges Fig. 14, all upper level principal tracks shown there can be found here, with the only exception being the southern branch over Eurasia. In this revision, the authors also showed results using 300 hPa height in the supplement, but those troughs/ridges are still absent at 300 hPa. In their response, the authors claimed that the waves in the southern branch are best seen in band pass filtered data, but in fact that's not the case. Chang and Yu (1999), and Chang (2005) both used unfiltered data, not band-pass filtered data as claimed by the authors in their response. Chang and Yu (1999) showed that these waves generally have longer periods (8 days or longer) than waves in the oceanic basins, and the composites shown in Chang (2005) also showed that these waves have low phase speed and longer periods. Hoskins and Hodges (2002) did use spatially filtered data (but not band-pass filtered data), but my guess is that since Hoskins and Hodges used vorticity at 250 hPa, even vorticity data that is not spatially filtered would show these systems. So the absence of these systems, even at 300 hPa, in the authors' results is still a mystery. In their response, the authors also stated that "This subcategory of waves however is still part of our data" and showed an example of that. The question then is why they don't show up in the climatology, while they showed up clearly in climatologies of previous studies (e.g. Fig. 12f and 13b of Hoskins and Hodges 2002)? The authors stated in their response that "In the revised manuscript, we placed a comment on this in the corresponding section". I might have missed that but I don't think I saw that comment in the revised manuscript. Can the authors point out where that comment is? In my opinion, as this seems to be the main difference between results of this study to those of previous studies for winter systems, this should definitely be pointed out and discussed in the paper.
2) p. 10, lines 16-17: Could that be related to the authors' choice of not including areas within closed contours as part of troughs?
3) p. 12, line 24: "bandpass-filtered eddies". This is also related to the authors' response to one of my previous comments. If I am not mistaken, Penny et al (2010) did not use bandpass-filtered data. They did remove the large spatial scale and long time scale (90-day high pass), but I don't think characterizing their data as bandpass-filtered is accurate.