Dear Franziska Frank at al.,

Thank you for yours answers and for preparing the revised version.

I must say that I am still puzzled by some of your findings.
Especially, I think you should address the critics of Reviewer #3
somewhat more who stressed the point that a lot of stratospheric
observations support gamma_H2O=2 and this ratio would break down above
the stratopause. Therefore I still have some questions regarding your
study:

major issues:
=============
To my understanding, there is no process above the "freeze-drying"
altitude, that would be able to change total water (H2 + H2O + 2*CH4).
Orders of magnitude are 4-7 ppmv H2O, 0-2 ppmv CH4, and 0.5 ppmv H2.
This is consistent with your figure 12.
The H2 mixing ratio throughout the troposphere and the stratosphere
is about constant at about 0.5 ppmv (likely due to approximate equally
strong production and loss processes).

Therefore, if the effective yield would be significantly below 2 and
H2 stays approximately constant at ~0.5 ppmv, the hydrogen atoms must
be found in a significant amount in a different (intermediate?)
species. This seems not the case in the results displayed in your
Figure 12.
Throughout the stratosphere (below ~0.2 hPa) the sum of H2O + 2*CH4 is
constant. For me that means that effectively all CH4 molecules are
converted into 2 H2O molecules (potentially involving equal production
and loss of H2).

In other words, if there were no H2 contribution from the troposphere,
you would see an increase in H2 with altitude due to CH4-oxidation
with gamma_H2O < 2. So one may need to consider gamma_H2O and
gamma_H2, but also the oxidation of H2 forming H2O.

The question is also, at what altitudes this is important. A deviation
from the factor 2 is clear above 0.2 hPa, but in the lower
stratosphere it is (at least) not important. This is so because
of the comparison between the altitude profiles of the
effective yield (Fig. 3) and the CH4 loss rate or the CH4 lifetime
(Fig. 4). If (in the lower stratosphere) the chemical lifetime is
above the typical transport times or in other words the loss rate is
very low, then the derived yield is not very relevant.

Could it be that your effective yield is below 2 because you do not
include all follow-up reactions of the CH4 oxidation? Or enough
"recycling cycles"?

Figure 10 shows the effective yield in the 3-D model. The values > 2
are said to be due to transport of intermediate species. It is not
clear, what these intermediate species would be? Is it H2? No other
intermediate species seems to have long lifetimes and/or a mixing
ratio above the ppb level.
Also that means that this yield would be dependent on the specific
transport formulation.

I appreciate that you included the subsection 4.1 for GCM recommendations.
However I fear there is not much use for GCM modellers. The idea of
a pressure-dependent gamma_H2O is likely not sufficient, as I think, it
should at least depend also on latitude.
It seems difficult to use a yield that is dependent on the specific
transport formulation. Would be better to involve the results from the
Lagrangian results that are displayed in figs 2 and 5 and a simplified
system involving CH4->H2, CH4->H2O and H2->H2O?
This would do the transport of theses threee species consistently with
the others.


Reviewer #1 asked you to include the results from the study by Wrotny
et al. (2010) who show a derived yield of 2.3 at 4.6 hPa while in your
Figure 10 the yield is about 1.8 at this pressure level. Could you
comment on that?


minor issues:
=============
(page and line numbers from the version with highlighted changes)

p.18 l.13. I think you cannot assume that the proportion of H / total
hydrogen is constant. H does have a clear diurnal cycle while total
hydrogen does not

figure 13. It is unclear what are you showing. Is it something like
d(H2O-tagged)/d CH4 and d(H + 2H2 + 2H2O)/d CH4 ?
please clarify.

discussion 3rd method with OH constrained:
If you use a constant OH value that may be inconsistent with the
remaining chemical composition, total hydrogen would be not
conserved. Is this a problem or do you see deviations in total
hydrogen itself in addition to the H2O yield?



typographical issues:
=====================
(page and line numbers from the version with highlighted changes)
p. 19 line 3: you likely mean "total hydrogen content"

fig 12, legend blue dashed should be "H + 3 umol mol^-1"
(alternatively you could use a x-range from -1 to 18 and
display H without shift)

Dear Franziska Frank at al.,

Thank you for yours answers and for preparing the revised version.
I am sorry, but I am still not convinced of the argument about the
transport of intermediates. Therefore please answer the following questions:

Looking on your figure 10 at 0.3 hPa there is yield of 2.2, that means
10% more H2O production compared with the CH4 loss rate. At this
altitude, I read from your figure 12 about 0.25 ppmv CH4, 0.5 ppmv H2
and 6.5 ppmv H2O, figure 4 gives a CH4 lifetime of ~0.8y.

Therefore, the CH4 loss is in the order of magnitude of 0.3 ppmv/y
and the imbalance must be about 0.03 ppmv/y. The shorter the lifetime
of an intermediate species, the larger mus be the concentration in
order to create this amount of imbalance by transport, i.e. 0.06 ppmv,
0.36 ppmv and 1 ppmv for a lifetime of 6 months, 1 month and 10 days,
respectively. Please clarify, what the intermediate species this would
be.

new figure 12b:
Why does tagged H2O have already about 2.6 ppmv H2O in the troposphere?
Does that mean, that already 1.3 ppmv of CH4 is already oxidized to H2O?

Dear Franziska Frank et al.,
Thank you for this clarification. With this I am happy to accept the paper for publication in ACP.
With best regards,
Jens-Uwe Grooß