Torner et al use a speleothem record from the Western Mediterranean region to provide new absolute age constraints to the glacial terminations of TIV, TIII and TIIIa. They use speleothem proxy data to provide additional hydroclimatic context to these time periods. And provide extensive discussion on how this record can be used

• to evaluate orbital forcing on Terminations,
• to evaluate the differences between millennial structure of the different Terminations,
• to provide absolute age control to marine records through tie points
• and to provide a fuller picture of the sequence of climate events during Terminations.

It would be great to see this article published. I have only a few comments that would hopefully make it easier to garner more information from the article.

Comments:

• In the introduction, please can you add just a sentence or so on why it is relevant to study Terminations in the context of current climate change
• Lines 65-70: It would be great if any speleothem could truly have an unambiguous climatic interpretation! It would be better to phrase this as ‘relatively unambiguous climate interpretation’ or ‘better understood climate interpretation’.
• Lines 105-110: Sub-heading 2.1 should be more appropriately named as maybe ‘ cave setting and regional hydroclimate’ or so. All the cave description could come first followed by a description of regional hydroclimate. So the sentences at Line 122, ‘The Murada cave….’ Onward should move to the start of that section.
• Section sub-heading 2.2 should be Geochemical analyses.
• Lines 140-145: Was the reversed age removed or included in the final age model generation by StalAge?
• Lines 140-145: Add MIS and Termination numbers to this figure to make it easier to understand the text around Lines 190. Please can you also show the exact sample points as markers either in this figure or in the appendix. It would be good to see how many samples cover the slow growth periods.
• Lines 150-155: ‘Concurrently measured’ should instead be ‘made on aliquots of the same powder sample’. Also remove the word estimated since these are simply results that have been measured and compared.
• Lines 165-170: What is CCiT-UB?
• Lines 265-270: Since so much of the discussion surrounds the sequence and duration of events, it would be great to have an idea of the uncertainties of these records, i.e. not just the speleothem record produced in this study, but also the other records being discussed in this figure. If the data are available, these could be plotted in the figure, if not, some estimates as a table or in the text would be great. Similarly, it would be good to see in the figure or have it noted in the text, if any event is represented by a single sample measurement. And the same for Figures 4 and 5.
• There are some typos in the text so it could do with another check.

Torner and co-authors present a new speleothem record from Iberia that covers glacial terminations IV, III and III.a. Because of the exceptionally precise dating of the speleothem, this record provides new age constraints on the timing of glacial terminations in the Western Mediterranean region. These geochemical records provide information on termination dynamics in response to forcings, as well as their duration and the respective regional climatic conditions. While the last two glacial terminations are increasingly well constrained through absolute chronologies, much less is known about previous deglaciations, and this record fills an important gap in our understanding of late Quaternary deglaciation dynamics. Thus, it addresses a very relevant scientific question that is well within the scope of Climate of the Past. The methods are state of the art and well carried out, and the discussion is detailed and not overstating the significance of the dataset. This is a very well written and well organised manuscript, and I have only minor comments that hopefully will make it more accessible to the readers, when published.

Minor comments:

- As far as I understand, this study includes new data from two speleothems, RAT and VALL2, and one marine foraminifera record. However, the manuscript only focuses on the RAT record, barely mentioning the other two datasets. I think it makes perfect sense for the study to focus on speleothem RAT, as the most complete and best dated record, but I think the manuscript would be clearer if the VALL2 and foraminifera records, their establishment, and provenance were described together with RAT in the methods, as well as mentioned in abstract and introduction (where the RAT record is mentioned). This would strengthen the manuscript as it would show that the results are based on more than a single speleothem record.

- line 30: use "northern" instead of north hemisphere

- line 67 and following: I find this sentence is worded a bit strangely, as it juxtaposes records with a specific climatic interpretation (Asian monsoon) to those from a region (America). Maybe consider adding the key climate patterns for American records as well, I assume they were left out because the references are indicating both North and South American records.

- line 83 and following: following my first comment above, I would mention the foram and the VALL2 records here as well.

- Material and methods: I would start this section with the description of both caves (Murada and El Pas de Vallgornera) and their respective speleothems, followed by the marine sediment core data. Then move on to the geochemical method descriptions. This would probably also help to streamline the manuscript a bit, since a lot of the methods are the same.

- line 149: "reversed axis" instead of "reversal axis"

- line 192 and following: I would join the two sentences ("According to the depth-age model..., while the warm marine isotopic stages..."). Also I think it's confusing to use different terms to describe opposing isotopic trends (here "high" and "light"). Please stick with one.

- line 196 and following: in the discussion on PCP mechanisms, I'm missing the potential influence of cave ventilation, which can drive PCP as well as drip rates. Is this not a factor at the studied caves or why was this mechanism omitted?

- line 207: "lower respiration rates", I would clarify that soil or ecosystem respiration rates are mean here, as it comes a bit abruptly.

- line 209: maybe clarify that the higher initial d13C at low soil pCO2 is because of the proportionally higher concentration of atmospheric CO2 (with more positive d13C) in the soil gas?

- Discussion around line 280: I wonder how comparable the magnitude and structures of the events shown really are, since these are all qualitative proxy records (and not all the same proxy either), and not quantitatively reflecting processes? I would phrase more conservatively, since we don't know exactly how sensitive the different proxies react to local drivers and the climatic perturbations they record.

- line 286: It is not clear to me what the influence of precession on Mediterranean oceanography is. I think it would be good to clarify with another sentence.

- line 392: Again, two different terms for isotopic trends are used ("light" and "negative"). I would stick to one and use it for both records.

- line 417: This sentence in the conclusions highlighted to me that the discussion did not highlight enough how large the discrepancies between tuned and absolute chronologies are. This is a really important finding and I think it should be brought out more in the discussion.

Figures:

- Figure 1: I find it difficult to understand from this figure and the caption which sites are the ones the data in this study comes from. Maybe add spell this detail out in the caption, and also consider different symbols or symbol sizes for the different "categories".

- Figure 3: In the top bar with the ages, the colours are quite difficult to distinguish, particularly Ejulve and Corchia. I would also consider making the symbols just slightly larger and the error bars slightly thicker (I know there is not a lot of space as the plot is dense) to improve readability. I think it could also be nice, but I understand if this is not possible due to the denseness of the plot, if the d18O and d13C records of RAT could be included.

- Figure 5: The line for the interglacial acme is difficult to see on the plot. In the caption, I would clarify "the d18O freshening structures in Iberia" (line 374).