Dear author,

Thank you for your responses to the two reviewers. It seems good.
I invite you to send the revised version of your manuscript with the amendments in track mode changes and after a look on the corrected version I will finalize the decision on your paper.

with my best regards

Nathalie Combourieu-Nebout

Dear authors,

Thank you for this revised version.
Your correction seem to be in accordance with the reviewer’s requests.

I have some final remarks on your document especially on the figures. There are some typing errors and precisions to do:

First, in all diagram is it possible to report the scale (cm and or cal yr) on the two sides (left and right or top and bottom), it will be easier to see correlations between the different curves. This has to be done on fig 3, 4, 8
- Figure 2: it remains very difficult to see the grey shadow. Perhaps with another color it will be more readable.
Figure 3: The problem of pine has been underline and you respond satisfactorily to the reviewer but I do not see the additional paragraph at the place indicated in your response. Probably it correspond to the paragraph between line 291 and 299. I understand that you interpret the high percentage of Pine as a migration of the tree line. But for me it is not sufficient for the third sample.
In fact, I have a doubt on the third sample as this sample is almost completely full of pine. Do you really think it is reliable? In fact it is very strange to have almost only Pine in a sample of a continental site. Does this sample not rather reflect a problem of differential grain conservation? If it is the case, this sample has to be removed from the diagram or at least it has to be more extensively discussed if you keep this sample because it remains highly questionable.
As the peak is defined and by one sample composed only by pine it has to be taken with cautious in my opinion.
The following peaks do not show the same picture and seems more reliable even if defined by one sample.

- Figure 6: I do not understand what are the small dendrograms in the upper right corner of each blue figure. In fact they are too little to be red. Remove them if it is not used in the paper or enlarge them to propose them in a readable way.
- Figure 7:
Decrease is not written in a good way (decrase) for forest, African dust and same for increase (icrease). Please verify all. The characters in yellow, light green and light orange are very difficult to read but perhaps it is a printing problem. Verify that please.
When you mark runoff decrease it correspond to an abrupt stop of the runoff at the beginning of the arrow. Wouldn't it be better to write stop and not decrease?

In the text, you added something on Olea but it is not clear for me. For you which pollen have the thicker endexine (and not intine which is not present in the past pollen) and which one has the higher size of reticulum? I think it is Olea and not Phyllirea. Please rephrase you sentence.
In the same paragraph, I think that precision are needed on the way you calculate the percentages. What does the basic pollen sum contain: all the grains or only the 300 without aquatic? Is pine included in it (probably)? You have to precise that. Please add something.

Could you please take into account these last remarks and amend your text accordingly. It is not a long time to do that. I am waiting after this new version before definitively accepted your paper.

Thank you in advance
Best regards

Nathalie

Dear authors,

Thank you for your response and last corrected version.
Your paper is now ready for publication.

Best regards

Nathalie Combourieu-Nebout