Dear Dr. Kotarba,

Thank you for your revised submission. After reading it through together with your Response to Reviewers, I feel you have addressed these comments sufficiently. I am also fine with adding the two new figures to a Supplement as you have done, rather than the main paper.

In my reading of this latest version I did note a few minor issues. I am happy to accept this study for publication in AMT once these are addressed, and don’t think another round of peer review will be needed. My comments are as follows:

Lines 44-45: as you note later in section 2.1 the Q value used to inform this categorical classification does in fact provide a quantitative confidence of cloud-free conditions. Although in Sections 2.3.2 and 3.2 of the ATBD (https://atmosphere-imager.gsfc.nasa.gov/sites/default/files/ModAtmo/MOD35_ATBD_Collection6_1.pdf ) the mapping is discussed in terms of “confidence”, elsewhere in the ATBD this is referred to as a “probability” of cloud-free conditions. I am not sure if this quantitative Q is provided in the product (I didn’t see it from a check through the file spec). But I think it’s worth amending the text here to note that the algorithm DOES calculate something which is interpretable as a quantitative probability, or to note (if Q isn’t in the files) that the only info provided to the user is the categorical mapping. I do agree with you that this doesn’t explicitly account for partial cloudiness within a single IFOV.

Section 2.3: here and in Section 5 it is worth noting briefly that since both Aqua and CALIPSO fly in formation in the A-Train, that matched IFOVs are from the same portion of the MODIS swath (i.e. same view zenith angle) throughout, not from across the whole swath. As cloud detection sensitivity is a function of geometry, it’s possible that the results reported here would differ quantitative for other portions of the MODIS granule. So the recalibration of cloud amount based on CALIOP matches may not hold for the broader MODIS data set. As one example of this you note you exclude MODIS Sun glint, although that this is one of the cases where MODIS cloud detection is more difficult. As glint is found in the tropics, the CALIOP-based cloud amount recalibration might not be as reflective of the true picture here since these cases aren’t found in the matched data set. This isn’t something which the data can tell us but should be mentioned as a caveat of the study. (I note that the sunglint bit is discussed again in Section 4, but the fact that one can’t test the sensitivity of recalibration to view zenith angle isn’t discussed, as far as I can see.)

Figures: the text and captions don’t directly state the gridding used here; this should be added to the text somewhere. I infer it is 2.5 degrees from line 295 but there should be a blanket statement about the gridding somewhere, for clarity.

Supplement: both Figures here are labelled S2, so the first needs renumbering. (This is about the Supplement file itself, not these figures appended to the main manuscript file.)

Please let me know if you have any questions; I think this study is a valuable contribution to the journal.

Best wishes,

Andrew

Dear Dr. Kotarba,

Thank you for making these further revisions; following a read through I am pleased to accept this version for publication in AMT.

I did notice that the Supplement now contains a duplicate of Figure S1 (it has Figures S1, S2, and then a duplicate of S1 also labelled S2). This can be corrected when uploading final files for production.

Best wishes,

Andrew