Dear Dr. Chevallier,

Firstly I would apologize for the long revision process. This is due to the difficulty in gathering reviews and the need to provide an Editor technical evaluation before publication you will find here.
I support publication for the paper, that is of definitely high quality, after addressing a final revision process that includes reviewer feedback and the Editor review that I may upload later on the ACP system.
Do not hesitate to come back to me for further clarifications.

It would be useful to make use of the detailed discussion provided in the response in the paper's body to provide more information to readers.

- Following reviewer 3 first two points authors correctly state in the abstract that the paper doesn´t provide claims on the usefulness of Satellite datasets in the retrieval. The answer to the reviewer is important and may be helpful to have a more through discussion to best reflect the strength of the analysis and avoid misunderstandings to the readers.
So, I would suggest to make use of the replies given in the Author's response on the discussion of the benefits of OCO vs GOSAT, the outcomes and the possible additional analysis needed – now in the conclusion – in the discussion section with also few additional details to discuss the coherence of Takagi et al results now added as reference.
The conclusions may then reflect the abstract statement.

- the same may be done for the use of aircraft data, e.g. with respect to the transport model uncertainty, and the coverage of aircraft and CONTRAIL measurements and their respective rms differences to provide such information in the paper text since reader's may not go to the discussion documents.

I hope you may be able to tackle these few technical rephrasing and provide a ready-for-publication version

Kind regards,

Federico Fierli