Dear authors,

I have checked your responses to the reviewers (both requested minor revisions) and found them over all satisfactory. However, I found two issues myself for you to consider:
1) In Figs 6 and 7 the error bars look symmetric around the mean although the values are plotted on a logarithmic scale. Is this by intention? This would mean that the errors on the left (negative) side are much smaller than those on the right (positive side). This is not consistent with the usual definition of the standard deviation. Could it be that the error bars you show are the standard deviation of log(delta_specific_humidity)? If so you need to clearly note that.
2) You mention that for finding collocations between the RO measurements and profiles from the model calculations, you calculate the minimum distance between latitude, longitude, and time of the RO observation and the respective latitude, longitude, and time of the model. How do you calculate this norm exactly, given that distance and time difference have different units? Please explain this in the text.
I'll be happy to accept your manuscript for final publication once these two issues will be clarified.

Kind regards,
Gabriele Stiller

Dear authors,

it is my pleasure to accept your paper for final publication in ACP.

Kind regards,
Gabriele Stiller