In this paper the authors seek to raise awareness of the contribution of external forcings, particularly historical, to climate model simulation uncertainties. Using the CESM large ensembles, including some new simulations using alternate CMIP forcings, they quantify the relative impacts of model changes and forcings in the CESM model framework. The newly presented CESM2-CMIP5 simulations are made publicly available as a resource for the wider research community to further investigate the impacts of forcings in the CESM large ensembles.

The paper is generally very well written, with clearly presented results supported by appropriate figures. The subject matter is a good fit for GMD and I recommend publications of this article subject to some minor changes.

Some particular issues to note are as follows:

1. One has to be very careful when talking about the direction of fluxes – particularly when considering net fluxes and differences therein. In that regard the flux definitions in section 4 need to be made a bit more clearly with mention that net fluxes are considered positive down (or “downward”).
2. Some of the ensemble naming is a little confusing. In particular it is a little confusing that “CESM2-LE” is used for the whole ensemble or for the 1^st 50 members. It would be best to be a bit clearer here.
3. The 15 CESM2-CMIP5 simulations are introduced as being “newly presented” but it looks like DeRepentigny et al. (2022) has already presented results from 10 of these simulations (or are they different?). This point should be made somewhere. The fact that some of the simulations might have been used already elsewhere does not take away from the novelty of this paper.

I include a pdf containing more detailed comments.

Please see the attached.