Interactive comment on “ Diagnosing the transition layer in the extratropical lowermost stratosphere using MLS O 3 and MOPITT CO analyses ”

Introduction Conclusions References

1.One major distinction between this work and the diagnostics shown in Pan et al., 2007 is that the comparisons are between model runs and observations in the latter but between the model with and without data assimilation in the former.Since the assimilation and analyses system in this case is also to be evaluated, and the satellite data involved have limited information and their own issues, the conclusion from the comparisons is not as straightforward.Physical arguments need to be invoked in each case to justify if assimilation/analyses result is better than the free running CTM.A major weakness in the paper is the lack of conclusive physical arguments in the discussions of the comparisons.
In Figure 3, for example, the data points from the selected vertical cross section in the model produce a "concave" (subject to the definition) fit after assimilations of MLS ozone.What is the physical meaning of the "concave" behavior in the tracer-tracer space versus the "convex" behavior in most previous observations?The "convex" behavior can be understood by the source reservoirs involved in mixing, as explained in Plumb (2007).What are the mechanisms for producing the "CONCAVE" behavior?Is this an effect sampling?nected to the tracer correlation plots.Do you imply an increase of ExTL depth in the geo-space?Or simply a wider scatter in the tracer-tracer space?If both, there need to be a rigorous connection between the two.
2. A significant change in the method of analysis made by the authors is to use 360 K surface, instead of the tropopause, as the reference point to construct the relative altitude distribution for the ExTL.The argument used for this change is that based on the inspection of figure 2, the 360 K isentrope appears to follow better the chemical tracer variability.There are a lot of problems with this line of reasoning: a. First, the hand-waving argument of 360 K is better is based on an inspection of a contour plot where the choice of color scale can change the perception entirely.The current color scale highlights the variability around 300 ppbv of ozone.The authors should try to see what the figure looks like when the scale is shifted to lower end, near the 100 ppbv of ozone.In any case a more quantitative analysis is required to support such choice.b.Second, the idea of reference surface is to choose a physical boundary and examine the tracer behavior relative to the physical boundary.The 360 K surface is not known as a physical boundary.What is the physical meaning of the relative coordinate profile in this case?c.This line of analysis leads directly to the problematic conclusion, largely drawn from Figure 5, where the two distributions, the ExTL points distribution and the tropopause height distribution, are compared in the 360 K relative coordinates.The better alignment of the two distributions in the 360 K relative coordinates does not support the conclusion of a better coincident (correlation) between ExTL and the Thermal tropopause.
3. The paper is titled ". . . the transition layer in the extra-tropical lowermost stratosphere".Similar wordings appeared a number of places in the paper.This wording shows a significant confusion and contradiction, since the "transition layer" is by definition between the stratosphere and the troposphere.C8758 4. The authors used the monthly average tracer fields in this analysis.The STE activities near the extratropical tropopause are largely associated with synoptic scale dynamical processes that are of a few days to a week time scale.The variability in tracer fields induced by these activities is largely smoothed out in the monthly mean fields.
Not as major but still important issues: 1.The information content of MLS O3 and MOPITT CO are very different due to their very different spatial sampling.MLS is a limb sounder and the O3 retrieval represents ∼ 3 km vertical layers that are several 100s km in horizontal scale.The MOPITT is a nadir instrument that has ∼ 20 km x 20 km foot prints but ∼ 2-3 pieces of information vertically.Its information content is largely concentrated in the mid troposphere ∼ 500 hPa and not very much near the tropopause.How would these sampling differences impact the assimilated field?This issue deserves some investigation and discussion.
2. The case study for the STE event on 15 August 2007 has been presented in detail in a previous publication (El Amraoui et al., 2010).The repetitions in this paper should be limited to a minimum level necessary for introducing the follow up work.

Minor and technical issues:
The labels for the figures are too small and very difficult to read.Interactive comment on Atmos.Chem.Phys. Discuss., 12, 22023, 2012.