Dear authors,

I am planning to send the paper back to the reviewers for their opinion on the revised version, but before that I would like you to more thoroughly address the major comment about the chl-a proxy for PMOM emissions addressed by referee #1. The new paragraph in the manuscript moves towards that direction, but you should be more specific and use more references there; you mentioned Long et al. (2014) and Rinaldi et al. (2013) in your reply, but have not used these two references in the newly added paragraph. One or two more references the reviewer suggested might also fit in the manuscript, and their value (or not) in the topic of your manuscript should be properly addressed. I completely agree with you that short-term campaigns might not be a good constraint for global flux parameterizations, but the same applies for the 8-day lag that was only based on North Atlantic data. In addition, this lag will not explain the fluxes over oligotrophic waters. I can only guess that the reason why Quinn et al. (2014) did not produce a parameterization is that the uncertainty of this whole system is so high, that any parameterization is prone to large errors, until we really understand the drivers of the organic enrichment much better, both in the field and in the lab. Please note that the reviewer uses the term “fundamental inconsistencies which have important implications for the major conclusions of this modeling study”; I do not take this statement lightly, so please try to be more detailed in your reply, in a way that will be also visible in the manuscript.

On another topic, I also want to add on Jeff Pierce’s comment that the use of all-sky AOD will likely lead to higher AOD values compared to the clear-sky AOD on average, since the relative humidity is higher in the vicinity of clouds, thus hygroscopic aerosols will grow more. As an example (no need to reference it, unless you find it necessary) you can see the differences between GISS_CS and GISS_AS results in: Kim, D., et al. (2014), Sources, sinks, and transatlantic transport of North African dust aerosol: A multimodel analysis and comparison with remote sensing data, J. Geophys. Res. Atmos., 119, doi:10.1002/2013JD021099. This means that your discussion on the topic should be mostly focusing on the possibility of an overestimation of AOD from your model, rather than a generic uncertainty discussion. A sentence on the direction your conclusions will move if the modeled AOD is overestimated will be very useful.

Thank you

Kostas Tsigaridis

Dear authors,

I have now received the answers from the reviewers. Two of them are very satisfied, but one believes that you had not adequately addressed their comments and serious concerns. I am attaching their detailed answer as a pdf file. Please try to either satisfy the reviewer, or document your disagreement with them appropriately in the manuscript.