|The revised version did improve from the ACPD paper, especially the|
attempt to include a comparison with other ozone loss estimates is
good, although this is a rather difficult task given different dates
and vortex edge definitions.
However, there are still some issues that should be clarified prior to
publication as explained below.
1. Validity of simulated transport, especially for passive ozone
More details of the transport simulation setup are presented in section
The transport is verified by comparison with N2O observations.
The inclusion of the vertical transport does of course improve the
model comparison, however, the difference e.g. of 20 ppbv on day 90 at
600K is probably caused probably by too strong mixing into the vortex.
This could be a source of error in the later results when using the
passive ozone tracer. Especially at latitudes below 65N where SMILES
observes mixing may even stronger as small scale filaments with high
PV may be represented worse by a too diffusive model. (this could be
checked e.g. by plotting a figure like fig.6 but with data restricted
to geographical latitudes below 65N).
[since I didn't bring this up clearly in the first review, I do
not insist on this point]
The derived ozone loss is calculated by subtracting passively
transported ozone from assimilated ozone. The value of passive ozone
very critically depends on the initialisation and the transport
characteristics of the model. Especially when only air-masses south of
65 degrees are used, that may contain high-PV vortex filaments in
which the transport and mixing characteristics of the model is
line 188: Does that mean that you initialized the passive ozone with
ECMWF ozone values? I thought it was initialized from assimilated
ozone at 1 December 2009. Please clarify.
Figure 5 differs slightly from the previous fig 4. Is it just some
other smoothing/ plotting routine or a different simulation? Please
correct also the panel titles of fig 5.
2. Definition of modified PV
line 194 modified PV
There are different definitions of modified PV in
the literature (Lait et al.; Mueller and Guenther, 2003; Manney et al.)
please specify method and reference level.
Changes in the results between averaging over the PV criterion rather than
EQL>70N suggests that the difference between EQL and PV criterion does
change significantly with altitude. Is this PV criterion meaningful for
the whole altitude range and how does it differ with respect to derived
At all locations in the paper where PV is mentioned, it should be made
clear, that modified PV is referred to.
3. Statements to PSC types
287ff: Even though it is not central to this paper, it has been shown
that the observed denitrification can indeed be explained without ice
formation on NAT (Grooss et al., 2014). There are more publications
that focus on NAT and ice formation (Engel et al., Hoyle et al.).
However, the "type of PSCs" is not really important in this context,
as already cold binary aerosols can activate chlorine (e.g. Drdla and
Thus, that I would suggest leaving out this sentence (also line 36/37).
4. Nighttime ClO
293ff: It is good, that the authors have incorporated this argument.
However, I would see the elevated nighttime ClO only as a sign of chlorine
activation. A quantitative eavluation would require modelling, but at least
information of zenith angle and temperature of each observation. I don't think
this should be done here. But without that, the number of globally averaged
ClO is not very meaningful and should probaly be left out.
l. 300 However...
This is formulated as if this would be a surprising new result to have
ozone loss by other mechanisms that chlorine chemistry. In the later
cited references refer to the NOx catalyzed ozone loss.
I suggest change the wording to something like
"The upper level (..) ozone loss is not due to chlorine chemistry..."
Also line 158ff could be read as if it would be a new finding that there
is descent in the polar vortex. Formulate rather as
"The descent of air in the polar vortex caused by radiative cooling
during polar night had not be taken into account in the previous model
version. It is, however, necessary for a correct evaluation of ozone
line 167: change omega to w
line 212: shown are N2O volume mixing ratios, not concentrations
line 227: capitalize Ozone
line 269: The second ozone loss period took...