Dear authors,

the two referees have now evaluated your revised manuscript.

While Referee #1 is nearly satisfied by your replies and just recommends some additional minor revisions, Referee #2 still recommends major revisions.

Both referees indicate displaying issues with equations and figures.
Referee #2 also points to more fundamental issues, including unclear methods, lack of scientific argumentation and quantitative discussion.

I believe the manuscript has a great potential, and I'm confident the group of co-authors can together easily address the above criticism.
I would recommend the authors to further review the manuscript, carefully taking into account Referee #2 criticism, and to make sure equations and figures display well.

Looking forward to receive the revised manuscript.


Best wishes

Dear authors,

based on the positive comments from one reviewer and my own reading, I believe the manuscript has significantly improved and it is now almost ready for publication.

Please, take into consideration the minor corrections from the reviewer, especially the one referring to line 447.

I also have few additional minor suggestions from my own reading (listed below) that you may want to consider before submitting the final version.
Please, return a change-tracked version for a speedy check.


Congratulations and looking forward to see this paper published on AMT.


Best wishes,

Nico Cimini



- Line 105: "atmospheric temperature and humidity content" -> "atmospheric temperature, humidity, and liquid water content"

- Lines 137-140: I suggest replacing the following sentence:
"is still limited, being a function of both radar frequency and atmospheric conditions (May and Wilczak, 1993). It is determined both by the attenuation of the sound, which is a function of atmospheric temperature, humidity, and frequency of the sound source,
and the advection of the propagating sound wave out of the radar’s field‐of‐view"
with:
"is limited by sound attenuation, which is a function of both radar frequency and atmospheric conditions (May and Wilczak, 1993) such as temperature, humidity, and the advection of the propagating sound wave out of the radar’s field‐of‐view"

- Lines 146: "has been" -> "has been and still is"

- Line 244: "inline" - not sure this is the proper word.

- Lines 303-307: I'm missing the introduction of the forward model for Tv and its Jacobian (at lines 361-365).

- Lines 320-323: I'm missing how sigma_Tb and sigma_Tv are estimated to be included in Se.

- Lines 328-333: I'm missing how LWP is estimated to be included in Sa.

- Line 382: I suggest replacing "to reduce the degrees of freedom associated with clouds" with "to eliminate uncertainties associated to clouds"

- Line 404: Maybe the authors mean the TOPROe background profiles from climatology? This is also noted by the reviewer. Please modify the text accordingly.

- Lines 714-720 (and in general): References to figure formatting (panel position, line color, marker type, etc) could be removed from main text and left in the figure caption only.

- Line 736: Any comment on the why RASS 449 Tv bias seems to depart above 1.3 km?

- Lines 741-742: I suggest adding the sentence:
"the RASS biases, because of the combined information from RASS and MWR."

- Line 836: "and above around 1.5 km AGL" -> “and only between 1.5 and 3 km AGL”

Dear authors,

thank you for providing additional clarifications.

I think the manuscript is almost ready for publication, but the following two comments that I believe should be addressed:

- Please, make explicit the number used for LWP uncertainty into the following sentence, as "large" is only relative: "LWP is arbitrarily assigned in Xa, with large values chosen for its uncertainty in Sa, so that it does not impact (constrain) the retrieval"

- Please, add the retrieval performances from non-BC MWRzo/MWRz to Figure 1A. As is, the manuscript gives no evidence that TROPoe BC provides better retrievals than original non-BC Tb. Actually, Fig.1A suggests that TROPoe BC provides no better retrievals than NN (based on non-BC Tb), specially in the boundary layer.
If the TROPoe BC does not improve performances significantly with respect to non-BC MWRzo/MWRz, that should be clearly stated.
In such a case, I'd suggest mentioning alternative BC sources to be considered, e.g. based on NWP profiles (as used in https://doi.org/10.5194/amt-13-6593-2020 and references therein).


Congratulations and looking forward to see this paper published on AMT.


Best wishes,

Nico Cimini

Dear authors,

thank you for providing PRs with no Tb BC.

One last remark on the sentence you added on line 852-855: it's true that the dashed lines (PRs without any Tb bias‐correction) show larger RMSE and bias in 0.5‐1.5 km AGL compared to TROPoe BC method, but the opposite happens in the 0-0.5 km; I think this should be stated.
Also, let me suggest to mention that the bias at lower V-band channels (shown in your Fig.1), likely causing the retrieval degradation above 3 km, may be related to larger absorption model uncertainty in that frequency range (Fig. 6 of Cimini et al, 2019 in your reference list).

I'm happy to accept the manuscript now, but would be grateful if you can address the above during the final submission.

Thanks much for handling the reviewing process with proper attitude.


Best wishes,

Nico Cimini