|This study employs idealized model experiments to analyze separately the impact of mesospheric and upper stratospheric ozone reductions as induced by EPP (direct and indirect effects, respectively) in boreal winter on the thermal structure and zonal winds in the middle atmosphere, as well as on surface temperatures. The topic of this paper is of high relevance since the inclusion of EPP as part of the solar forcing in the upcoming CMIP6 model experiments has recently been recommended. The choice of an idealized ozone forcing, allowing to separate the impacts of mesospheric and stratospheric ozone loss, is, in principle, a justified approach to investigate the different roles of mesospheric direct EPP impacts and stratospheric indirect effects and their underlying mechanisms, a topic that is recently widely discussed in the community.|
The authors have done in general a great job in responding to the comments raised by previous reviewers. However, the suitability and limitations of the idealized experimental setup for a quantitative assessment of EPP climate impacts is to my opinion still not sufficiently discussed. The authors have chosen intentionally a simplistic experimental setup in order to be able to specifically address the different processes related to direct and indirect EPP impacts and the identification of mechanisms for possible climate responses. However, this idealized approach is less suitable for a quantitative assessment of EPP climate impacts which would clearly benefit from the use of a realistic, transient forcing in consonance with recent observations. In this sense, the focus of the paper should be clearly on the former, and this should be better reflected in the title, abstract and the conclusions. As a minimum, it should be mentioned in the title and abstract that *idealized model experiments* have been conducted to identify the climate impact of mesospheric and stratospheric ozone loss *as caused* by EPP. It should also be clearly stated in the tile and in the abstract that the study focuses on boreal winter.
Some discussion on the limitations of the chosen experimental setup has been included in the revised manuscript. However, the limitations related to the use of an ozone forcing being constant in time - to my opinion the most relevant limitation - is mentioned only marginally. EPP impacts, particularly due to the indirect effect, are mostly restricted to polar winter and it is not clear at all how the application of an unrealistic ozone forcing in polar summer could interfere with early winter EPP effects. The polar summer temperature and zonal wind responses are large (see Fig 3) and lead to a strengthening of the early polar winter vortex that could modify the early winter EPP response. The authors speculate at p4 l18 (of the track change version) that "it is unlikely that signals in summer affect the climate of the next winter". However, more evidence needs to be provided to support this statement. It is further not clear to me at all what is the advantage of using an ozone forcing constant in time instead of using an annually repeating pattern. The choice of the former clearly needs to be motivated.
Overall, I think that this paper is well suited for publication in ACP after addressing the concerns raised above, as well as some specific and minor comments listed below .
Specific and minor comments (pages and line numbering refer to the track change manuscript):
p2 l12: please use Ap instead of AP
p3 l31-32: Here it is stated that ozone concentrations are averaged over 1850-1860, while in the response to reviewer#2 you say that you used "ozone profiles averaged over the late 20th century provided by
the general circulation and chemistry model HAMMONIA". Could you please clarify?
p6ff l32-l1: I wouldn't say that a heating response of 0.1-0.3 K/day is small. This is of the same order as what is caused by UV-induced ozone increases in the tropics.This is important since, at the end, it is the latitudinal gradient that is thought to be responsible for dynamical responses to both EPP and UV, no matter if it is introduced by a warming in the tropics or at the poles.
p13 l1-3: A possible reason for the different temperature response compared to the DJF responses in previous model studies could be the use of an ozone forcing constant in time. A more realistic forcing as used in the previous studies would result in a negligible early winter (heating) response as it takes until mid winter to bring ozone-depleting NOx into the stratosphere. This should be discussed.
p13 l13-21: The authors have added a paragraph that stresses the need for more research on the effects of EPP and its climate impact. This definitely a good point. However, they removed the sentence about the possible limitations of their analysis regarding climate impacts due to the choice of a simplified experimental design. Why?
Fig 4: The caption is not consistent with the figure of the left panel (showing absolute NH surface temperatures in DJF rather than SON SH responses).