Dear Dr Sabatino,

Thank you very much for your revised submission. One of the reviewers is pleased with the modifications you made. The other one did not react to my invitation to review the revised manuscript, so I made an assessment of your changes in the light of his comments. To my judgement, your reactions are adequate, and in principle the parer can be accepted. I noted, however, a few shortcomings (mainly technical) that you should address.

- For several Figures (8/9 [actually 9/10], 14/15, 18/19) you note that the color scales of panels a,b,e,f as well as c and d are now the same. Unfortunately, this is not true. E.g., in Fig. 9 the color bar goes from 0 to 4.5 in panels a and b, from 0 to 4 in e, and from 0 to 3.5 in f.
- In the paper and especially in the figures you refer to the observed data as "experimental" data. This is confusing, as in modelling studies "experimental" usually refers to "from a numerical run in which one or more parameters were changed" -> thus "experiment". So if possible you should replace "experimental" by "observation".
- From the description of the models (secs. 2.2/2.3) you removed the values of the constants and transferred their discussion to secs. 3.1/3.2. That's fine, but the reader is now asking "And what is the values of this constant?" To avoid confusion, you should mention in sec. 2.2 that the choice f the parameter values is discussed in secs. 3.1/3.2.
- sec. 3.5: You just state the differences in model output. A bit of interpretation to help the reader comprehend the the differences would be welcome. Are these differences large/small, (un-)understandable, (un-)expected, in line or not with other results?
- p 9, l 254: "for the 2007" -> "for the year 2007", or "for 2007" (occurs more often, please check)
- p 10, l 276: how -> that

Looking forward to your final manuscript,
Andreas Sterl