Summary:

This paper provides a summary of a number of experiments conducted at the Forschungszentrum Julich (FZJ) atmospheric profile simulation chamber designed to connect the ozone measurements profile measurements made as part of the In-service Aircraft for a Global Observation System (IAGOS) to those made on balloon sondes using the common, world-standard UV calibration instrument of Proffitt et al. (1982).  As such this paper is an important contribution to the literature and will allow a harmonization of in situ ozone profiles across these measurement platforms.

In particular, the experiments conducted examined the performance of two versions of the aircraft O3 instruments (P1-O3 and CAR-O3) against the dual-beam UV-Ozone Photometer (OPM) of the World Calibration Center of Ozone Sondes (WCCOS) at FZJ.  The instruments generally showed agreement to within 5-6% over the range of pressures studied.  Interestingly, the P1-O3 instrument showed a consistent trend in offset from the OPM, starting at ~+2% at 1000 hPa and changing linearly to ~ -3% by 400 hPa.  The paper was uncertain as to the cause, which does need to be identified and reconciled.  It mentions that the performance of this instrument might be an artifact of the experimental set-up.  That question should be resolved.

Recommendation:

Publish with relatively minor revisions – see my detailed comments below.

While it would be good to have the question of the drift in the offset of the P1-O3 instrument from the WPM resolved, it is worth getting these results into the literature sooner than later.  If it is not resolved by the time of publication of this manuscript, a follow-up “note” should be submitted with an answer to the question.

Detailed comments can be found in the attached supplement file.

As mentioned several times in the text this intercomparison study is a first step of the long-term goal to get the global ozone sonde data and IAGOS-O3 data sets traceable to one common reference.  This long-term goal is extremely important for the ozone research community and therefore this study is very welcome!  Accordingly, I recommend publication after addressing two major and the minor comments below.

Major

1. As mentioned in the text twice there is no ozone reference instrument running at reduced pressures at any National Metrological Institute in the world.  The OPM at Julich acts as a workaround to this situation.  However, we need such a reference in addition to studies like this one.  Although this gap in the whole concept is mentioned in the manuscript, I recommend to address it even more prominent and pronounced eg. in the abstract.  This would enable relevant persons better to ask for closing this gap.

2. The display and description of the figures can be crucially improved.  Each single issue is a minor one (and are listed in the minor section below).  However, overall, these issues make it sometimes hard to follow the arguments of the manuscript.  

Minor and typos

Figure 2:

There is no temperature data and the y-axis "Temperature" text can be omitted.  Please arrange the figure similar to Figs. 4b, 6, 7b, 9, and 11, ie. separate upper and lower panel.  Here and in other figures, please use lower case letters for ppbv.  There is no need to write VMR in addition to "Volume Mixing Ratio" at the y-axis.  For this type of figure, it would be nice to use both sides of the frame for y-axes descriptions instead of an additional y-axis on the left side.

Figure 3:

Some blue dots are masked by color descriptions.  Please avoid that.  Less cryptic axes descriptions would be welcome, eg. "reference pressure" instead of "p-ref.", "pressure difference" instead of "Delta p".  At least explain abbreviations if used in the plots, eg. "Delta p", "cuv.", "ref.", in the caption.

Figure 4 (& 7):

Panel letter descriptions, a b c, should be inside the plots.  4a and 4c: Separation into two parts like in 4b with help lines at -5, 0, and 5% would be helpful.

line 38: Please add a comma "... of the atmosphere, and its impact ..."

l 69: The information in the two brackets can be combined.

l 148: "... to prove that the linearity of the instrument is within 1%."

l 151: Unsure, what the word "above" should tell.

l 152: Is the "s" in "P1s" correct?

l 287-292: In the description of what can be seen in this figure, please add a comment on missing data, ie. absence of CAR-O3 measurements during about 30 min, partly during descent.  Any additional comment, what had happened at 14:40 UTC?

l 304: My understanding of the following text is, that the issue found with the electronic analog-digital converter was systematic present in all CAR-O3 instruments and not only in the instrument used in this study.  Please mention that explicitly here.

l 320-321: The statement "... occurs identically during ascent and descent and no indication for any hysteresis effects could be observe" is hard to verify with the current set of figures.  I miss an additional corresponding figure overlaying ascent and descent branches with eg. different colors.

l 420-424: Please swap the sequence of both sentences.

l 421-422: The relative differences for larger parts, eg. zero ozone periods, are not shown in the upper panel of figure 9.  Why?

l 421/422 & 431/432: The figure numbering seems to be odd.  The referenced figure 8 deals with experiment 4 and not with experiment 7.  A corresponding figure for experiment 7 is missing.  Please add such a figure.  Instead figure 9 figure 10 should be referenced.

l 427: Small differences of -(1-2)% for the P1-O3/OPM are mentioned.  I see values between -7 and +3 %.  Have I misunderstood something?

Table 3: In figure 10 one can see only the fits per instrument for the upper O3-ranges but not for the lower O3-ranges.  Another figure filling that gap would be helpful to verify such values like 0.942 for the P1-O3/OPM at 400 hPa / 0-200 ppbv.

l 448: malfunction

List of Acronyms: Please add RAM.

l 583: Staehelin

General: It would be nice to have somewhere an exact definition of the relative difference used, eg. = (O3_IAGOS - O3_OPM) / O3_OPM