I enjoyed reading Oldeman et al manuscript on the relationship between North Pacific climate and ENSO variability during the mid-Pliocene. The manuscript is well-written, and the figures are very clear. I had some minor edits in the .pdf version of the manuscript. I’d encourage the authors to explore the residual part of the Aleutian low (AL) variability a little more in-depth. If the AL variability is not linked to Arctic variability, what about the variability of western Pacific moist convection, and the resulted upper-tropospherical heating? It was suggested at various places in the manuscript that the convective heating and the generated Rossby wave, i.e. the atmospheric bridge, is the mechanistic link between the tropical and AL variability, but was mainly discussed in the context of ENSO. Yet, the variability of western Pacific moist convection probably won’t be entirely explained by ENSO, right? Can residual western Pacific variability explain the residual AL variability?

I find the manuscript well written and the scientific arguments are sound and well presented. Overall, this work qualifies for publication and I suggest here a few points that could improve the quality and description even further.

Specific points:

Line 40: As it has been argued that the mid-Pliocene is an analogue of the near future climate, it needs to be clarified here that the behavior of the AMOC is different in present to future climate in the climate model simulations (a declining AMOC) compared to what the PlioMIP2 models are showing (intensified AMOC). AMOC plays a key role in our climate system and therefore its direction of change under enhanced CO₂ is crucial.

Line 69-70: do you mean here that the ENSO variability change in future is different from the one we find in PlioMIP2 simulations, where it’s seen to be decreasing? Again, later in the paragraph at line 74-75 you mention that at high CO₂ forcing a weakening of ENSO variability is found. How do you reconcile these two parts? Why do they reach different conclusions. Could you give some hypothesis here? And how this present study helps in this context. Overall, I think this paragraph needs some more thought and work to make it not confusing and clearer.

Line 103: the term ‘sufficient amount’ sounds vague to be in a scientific journal. If on average the range of number of years can be specified that would be better.

Line 126-127: I wondered if the author needs PDO to be the part of this manuscript as the PDO and its connection have rarely been explored in this study and at the end, author’s schematic shows that there is another work focusing on PDO that is in prep. I would recommend that author can leave the PDO totally out of this paper. It’s not at all required for the point that the author makes in this paper.

Line 239: In the entire paper, I got confused between the terms multi-model-mean and ensemble mean which I think are used in the similar meaning. I would recommend to stick to one term. Either multi-model-mean (which I would prefer) or ensemble mean (I would not prefer this as it usually denotes means of multiple ensemble members of a single model and doesn’t really clarify that multiple models are involved in the construction of the mean).

Line 254: I think such correlation is just happening by construct. It’s the part of the AL variability related to ENSO to begin with. Therefore, by construct they are supposed to be having high correlation. That needs to be mentioned too.

Line 304: Typo- it should be ‘on the one hand’

Line 317-318: Framing some questions at the end of introduction and then answering them in discussion or conclusion is understandable but framing a new question here in the discussion for the readers seemed a bit odd to me. Please frame it as a topic to be explored further or a question that yet to addressed.

Line 333-334: need a reference here regarding the statement made here.