Many thanks to the authors for providing detailed responses to the comments of the three reviewers. The constructive comments of the reviewers make clear the value of the work presented here, but it is also clear that there are some misgivings about the discussion of the age model and its uncertainties. Although the details are available now across several previously published papers (as detailed by the original manuscript and author replies), the focus of the manuscript on the timing of events and on leads/lags means that careful consideration of the age control is vital. The authors replies are already valuable, but I also agree with the reviewers that this information needs to be incorporated into a revised version of the manuscript, so that readers are not required to dig into the different cited references to understand the evolution and nuances of the age control. There are serious concerns expressed by Reviewer#1 regarding the dinoflagellate cyst constraints, and these should be addressed carefully in any revisions (perhaps figure 1 needs to indicate the uncertainty around the dates given for these events?).

Given the focus on the timings of events on G-IG timescales, it is also important that the age control is clear, because the relative timing to the suggested forcings will also be determined by these uncertainties. Although the authors argue that several of their conclusions are about relative phasing between co-registered proxies, the abstract indicates a key finding related to obliquity pacing, which necessitates confidence in the absolute age scale as well. I agree with Reviewer#3 that an outline of how obliquity forcing might be translated into the different environmental signals recorded here could be valuable. If the authors can tackle the constructive suggestions made by the reviewers, then these claims can be better evaluated by the reader.

Reviewer#3 also makes a perceptive comment (page C2) about the potential for the leads/lags interpretations to have been affected by sample resolution and sample depth being different between proxies. I didn’t see a reply to this point by the authors, but this potential uncertainty ought to be acknowledged. The authors might consider making a detailed plot of the key G-IG cycles they discuss, along the lines of the one that Reviewer#3 generated by cutting/pasting the main figure from the text.

Reviewer#3 indicates that the methods section may become very long and dominate the manuscript, and they suggest putting some of this information into the supplemental. I leave this decision with the authors (it was not raised by reviewers #1 or #2), recognising that sometimes in a large multi-proxy study it is also necessary to have a long methods section.

The authors are requested to incorporate the changes that they have indicated they will make into a revised manuscript in order to address the constructive comments of the three reviewers.

Thank you for providing your detailed responses to the constructive comments of the reviewers, and for making a number of changes to the text and supporting materials which have strengthened your arguments. There are some minor typos or clarifications which I have noted separately but these are easily corrected.
Thank you for submitting your work to Climate of the Past.