Dear authors,

Your revised manuscript addresses the main points raised by the reviewers, and is therefore acceptable for publication in AMT. However, the manuscript is very lengthy, and I find it tough to go through, even while being familiar with the topic. Therefore I strongly recommend that you shorten the manuscript wherever possible. There is quite a bit of repetition in the text, for example when you extensively quote the outcome (including uncertainties) for all the comparisons, when these can also be seen from the tables or figures. The text reaches a total of almost 1000 lines, and I think this can and should be reduced by at least 20%, so that readers can more quickly digest the essence of your work. The work basically does not justify 1000 lines. One of the reviewers, while positive about your revision recommended a useful opportunity to make the paper more concise: "I’d suggest the authors combine the discussions of biases and uncertainty, so that readers could have a wholistic picture about the performance of each satellite retrieval".

One other reviewer had two useful comments, which should also be addressed:
(1) The abstract focuses very much on absolute bias numbers while the analysis (e.g. the regression slopes far from unity) indicate that these have limited meaning (many of the effects being multiplicative). Perhaps consider reporting %-biases or slopes (instead or additionally)?

(2) Section 2.4: Agreed on the importance of the assumed vertical profile in the satellite retrievals, and the value of homogenizing these, but what about the profile used in the retrievals of the GeoTASO and GCAS tropospheric columns? In 3.3, you write "GeoTASO and
GCAS retrievals were not reprocessed in order to have a consistent reference data set for satellite evaluation." I do not understand that argument. Why would reprocessing them mean they are no longer consistent (with what?)? There would of course be a concern about independence, but I believe it would have been a meaningful exercise nonetheless, unless I miss a strong argument against it. Actually, unless I missed it, little detail is given about the profile that underpins the airborne retrievals. Some text on this would be appreciated.

Specific comments from the editor:
Throughout the manuscript you use two words for many concepts, where only one should be used. Example: on line 30 you state "clean/background". Just use "clean". There many such cases.

L34: this sentence should be phrased such it makes clear that the number is a campaign average NO2 uncertainty. Now it could be interpreted as a generic finding, which is not the case. Same for the HCHO uncertainty on L42-43.
The abstract is too long - I suggest to shorten it by 30%.

L141: the equator overpass time of OMI is 13:30 hrs.

L167: please doublecheck that the NASA NO2 algorithm uses the OMI O2-O2 cloud parameters. They have been using the Raman cloud product from OMI (Joiner-papers) in the past, and I doubt that this has changed. Regardless I think the O2-O2 product citation should be Acarreta (2002) or Veefkind (2016), not Vassilkov.

L224: it is useful to remind the scientific community that v2.3 of TROPOMI NO2 has indeed improved compared to previous versions. The main reason for this is -better cloud pressures via FRESCO+ wide- explained and demonstrated in Riess et al. (2022), and it would be appropriate to cite that study here (along with van Geffen et al. (2022)) here.

L341: qa_values should not be zero or < treshold, but > treshold. Please correct

L360-367: these lines can be removed

L442-444: here the main reason why v2.3 has improved could be brought up as explained in Riess et al. (2022)

L445: that NO2 (and all spaceborne tropospheric DOAS retrievals) have a magnitude dependence is well-known since the early works of Martin et al (2002) and Boersma et al. (2004). Those studies showed that the uncertainties in the AMF drive this magnitude dependency.

Caption Figure 6: is this Figure showing the FNR derived from the NASA scaled or NASA standard products? This should be clarified in the caption.

References
Riess, T. C. V. W., Boersma, K. F., van Vliet, J., Peters, W., Sneep, M., Eskes, H., and van Geffen, J.: Improved monitoring of shipping NO2 with TROPOMI: decreasing NOx emissions in European seas during the COVID-19 pandemic, Atmos. Meas. Tech., 15, 1415–1438, https://doi.org/10.5194/amt-15-1415-2022, 2022.

Please go through the manuscript one more time, address the above comments, and make the paper more concise, and then resubmit. Thanks for an informative paper.

Dear authors - congratulations on a very robust and interesting paper. I trust that this more concise version will be well-read. Kind regards - the associate editor.