The authors analyze mid Holocene climate model simulations with and without taking into account the effects of prescribed Saharan greening (SG). The model experiments using EC-Earth have been described elsewhere, and the new aspect of this paper is the focus on mid-latitude atmospheric circulation patterns and changes in weather regimes (WR). The authors discriminate between changes induced by the orbital  and GHG forcing as in the standard PMIP3 simulation, and additional modifications of climate features by including prescribed SG.

The topic of the manuscript addresses a relevant scientific question in the discussion of causes for past climate changes on the regional and continental level, where a lot of discrepancy remains between proxy reconstructions and model simulations.

Extending the analyses to the mid-latitudes is therefore timely and useful and provides important insights.

The study underlines the importance of including changes in vegetation and the related dust fields not only for the monsoon systems, but also for mid-latitude circulation and circulation patterns. The authors find significant effects coming from SG for temperature and precipitation fields. Regarding variability patterns, the effect of SG is merely an enhancement of effects that are already seen in the standard PMIP experiments, although they are not making an attempt to explain why the SG effects would do that.

Another important finding is that one has to include variability on the daily timescale to refine the regional effecst by means of WRs, which sometimes give different patterns than circulation patterns based on monthly data. The authors also discuss if and how much the proxy-data gap can be narrowed down in the SG simulation, but with rather limited success. 

Overall, the manuscript is written in a clear and concise way and the findings are well supported by the prescribed analyses and figures. 

Save for a few inconsistencies, I find the manuscript quite mature and I would recommend that the paper can be published after (very) minor revisions.

Minor issues:

Abstract, ln 21ff: Don’t you say in the paper that the NAO+/- shift comes about in the run without SG and that SG just enhances this effect (c.f., Fig. 6) ?

Figure 1: blue dots, in particular over southern Europe are hard to distinguish from the gray ones, maybe make them slightly bigger?

Ln 99: how uncertain is this estimate?

Ln 115: a few more words about the Cohen and how it is applied over the selected regions should be included so readers don’t have to look it up.

Ln 295: maybe include also a discussion how that relates to the suggestions by Mauri et al. (Clim Past, 2014)

Ln 381: “impacts” of what?

Ln 394: what gives you the idea of looking into changing ocean circulation, any references for that?

Summary

The author analyse two mid-Holocene climate model simulations using EC-Earth. The first of these is a stnadard PMIP type setup and the second includes a reduction in Saharan dust along with a prescribed greening of the Sahara. The authors focus on mid-latitude climatic impacts and in particular the changes in weather regimes and the NAO. The authors also evaluate the simulations with proxy records for temperature and precipitation. They find significant but similar impacts on the NAO and WR independent of the greening, but the agreement with proxy records is worsened in all areas but one when the Sahara is greened.

Recommendation

My only main comment is that some of the key findings could be more clearly articulated in the abstract and the conclusions. I also have a few minor suggestions about clarity that are listed below. Otherwise, this is a valuable study and I look forward to seeing it published.

Main comments

A key finding here is that the MH_PMIP simulation performs better than the MH_GS in all regions but one according to the Cohen Kappa scoring. I think this needs to be clearly stated in the abstract. 

This slightly surprising finding could also benefit from further discussion - at the moment the Conclusions seem to argue for more realistic mid-Holocene simulations, but clearly there are nuances here. What could be the cause of this? One question that came up on reading was whether the highly idealised nature of the GS simulation setup could play a role?

Minor Comments

In the discussion of mid-latitude temperature change it would be worth referring to Bartlein et al (2017) who looked at this issue in multiple models.

Figures: There are a lot of figures here and the reader has to jump between the main text and the appendix figures quite a lot. Could you consider moving one or more of these figures into the main text to reduce this. At least figure A2 would be better in the main text.

Lines 45-47: "However, the interpretation of these climatic changes, particularly on temperature and precipitation patterns, as indicated by proxies, seem potentially inconsistent with the suggested changes in the atmospheric circulation (e.g., the positive-to-negative shift in the NAO/AO phase).” 

Perhaps you could elaborate on this in slightly more detail?

Line 231: “However, the difference in the NAOI distributions between the MHGS and MHPMIP experiments is less significant (p<0.11)”.

Do you mean not significant or just less ?

Lines 232-237: “Circulation and surface anomaly patterns associated with the NAO positive phase in the MHPMIP (not shown) and the MHGS experiments (Fig. 7c, d) are very similar.  …. In particular, the thermal and rainfall anomalies are more pronounced when Saharan greening is taken into account, due to the significant difference in the NAO phase shift with respect to the PI period.”

These two statements seem contradictory to me. Please could you clarify?

Line 287: “… suggesting that the effect of the Saharan greening on the atmospheric circulation and the associated thermal and rainfall anomalies amplifies the changes driven solely by the orbital forcing”

Can you provide any potential explanations for this?

Lines 367-369: “This suggests that the MHGS simulation more effectively captures precipitation patterns compared to seasonal temperature patterns across the mid-latitudes”

Shouldn’t this be MH_PMIP as the MH_GS runs has lower scores for every region except Asia?

Line 361: “numerous inconclusive MH proxy records”

Could you define what you mean by inconclusive?

Line 405: “In addition, it is shown that the simulated Saharan greening drives”

Shouldn’t this be the mid-Holocene orbit as the modes are similar in both GS and PMIP?

Line 409-410: “the changes driven by the Saharan greening in large-scale circulation indicate plausible explanations for the proxy evidence.”  

It’s not clear what this really means but it sounds like its contradicting the Cohen’s Kappa scores which show that the GS simulation is mostly worse than the PMIP simulation?

Line 412-414: “Furthermore, this modelling exercise also highlights the need for more refined MH climate modelling, such as prescribing realistic vegetation across latitudes and considering the seasonal vegetation cycle, to account for large- and local-scale climate feedbacks.”

These sound like sensible suggestions but they come slightly out of the blue here right at the end. Where do these ideas come from? Could you link them to your work or other studies in some way?

Technical Corrections

Hermann et al (2018) is missing from the reference list?

Figures 2 and 3: could you make the lines slightly less thick - as they are they overlap a bit too much.

Contributions: “QZ run the simulations” -> “QZ ran the simulations”.