Validation of Water Vapor Vertical Distributions Retrieved from MAX-DOAS over Beijing, China

Round 1

Reviewer 1 Report

please see attached review

Comments for author File: Comments.docx

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript validates the use of the ground-based MAX-DOAS method to determine the vertical profile, total column amount and surface water vapor content. The study is useful, its results will expand the number of used sources of data on the content of water vapor in the atmosphere. Information about water vapor is needed for a variety of purposes, which adds value to the manuscript.
At the same time, we have a number of remarks regarding the error budget and the comparison of the obtained profiles with independent data.
So, it is not clear why, when analyzing  the errors budget of the vertical profile (lines 355-364), only smoothing error and noise error are considered; it is also not indicated on the basis of which real covariance matrix the smoothing error was calculated. We would like to see a complete error budget analysis that takes into account Rogers' approach (see Rogers, 2000, “Inverse Methods …”  https://library.wmo.int/doc_num.php?explnum_id=1654, i.e. Eq. 3.16). At the same time, the analysis of errors of the general content (lines 368-380) is carried out in more detail. It is not clear why the same sources of errors were not considered for profiles. In addition, the authors do not at all distinguish between systematic and random components of errors, which, in our opinion, needs to be done.
The correlation analysis of the comparison of data from different measurements performed in Section 4 by the authors only partially characterizes the discrepancy between the results. In addition to it, the bias and SD of data differences are important, they should be calculated and given in the manuscript. In particular, Figure 9, which shows 20 (!) scatter plots, is difficult to understand and does not carry useful information. Instead, it is better to show the vertical becheviour of the random and systematic components of the difference between data from different sources, and compare them with the errors of the two types of measurements.
In the “introduction” section, the authors mention that the presented method was previously analyzed in [31]. It would be nice to see in the “Summary and conclusions” section how the results of this work compare with the new results.

A few notes on the text:

Line 182-183 and below “a priori error” - more correct “a priori uncertainty”
Line 313 “the bottom panel” - in which figure? The above mentioned was Figure S9!
Line 314 What “the statistical uncertainty” means here? SD? confidence interval? something else?
Line 355 “smoothing error, noice error and the sum of it”: Rodgers (2000) considers 4 components of the measurement error: the smoothing error, model parameter error, forward model error, and the retrieval noise. To estimate the smoothing error, it is necessary to have real covariance matrices of the gas vertical profiles.
Line 375, 376 “uncertainty related to the H2O cross sections” - it is systematic uncertainty, I guess?

Author Response

Please see the attachment.

Author Response File: Author Response.docx