Dear Dr. Rosselli,

Thank you for submitting your manuscript "Determining the hierarchical order by which the variables of sampling season, dust outbreaks occurrence, and sampling location, can shape the airborne bacterial communities in the Mediterranean basin" to Biogeosciences. After reviewing the comments made by the four referees and your response letters, I find that your manuscript requires major revisions to address the main concerns of the referees. Your overall responses to the referee comments appear very detailed and complete to me. Please incorporate your replies as much as possible into the revised version of the manuscript addressing especially the concerns of referee #2, #3 and #4. These include:

- Measures taken to prevent batch effects between different samples
- Removal of the term “ seasonality”
- Clarification on dusty vs. clean conditions
- Improvement of the method description and clarification of misleading parts (e.g. control samples)
- Improvement of Tables and Figures
- Clarify the relation to previous studies of the same authors
- Restructuring of the results section

Please note that your manuscript will be re-reviewed by at least one referee.

Best regards,

Ivonne Trebs

Dear Dr. Rosselli,

thank you very much for submitting a revised version of your manuscript. Referee #3 still has serious concerns about the very low number of samples (lack of replicates) used in this study, in-light of the three different variables that are considered (location, month, dust/clear conditions). The adivse is to compare two variables at most. Using ANOVA as a statistical test for differential abundance is an inadequate statistical approach for this type of data. Please have a look at the detailed report.

These concerns must be addressed in detail before the paper can be accepted in BG.

Best regards

Ivonne Trebs

Dear authors,

thank you very much for the responses and rebuttal of the referee comments. Before publication of your final manuscript, I recommend to add some more details (2-3 sentences) of your explanations given in the response letter about the statistical approaches (in particular Aitchison’s centered log-ratio transformation). Please also check that the letter size in your Figure files is large enough (also in the Supplement).

Best regards,

Ivonne Trebs