The manuscript, "Comparison of ocean heat content from two

eddy-resolving hindcast simulations with OFES1 and OFES2", by Fanglou

Liao, Hua Wang, and Zhiqiang Liu, suggests that OFES 2 does not warm

in accordance with upper-ocean observations and the previous

reconstruction of OFES 1 from 1990-present. The decomposition of ocean

heat content changes into heaving and non-heaving contributions is

interesting, but it doesn't conclusively explain what is going on with

OFES 2. The manuscript would be improved if it could both document and

interpret the dynamical causes of the discrepancy between OFES 1

and 2.

Major issues:

It is interesting that the ocean heat content changes primarily by the

change in isopycnal depth. Does the total heat content calculation

depend upon the calculation of heat content change by heaving motions?

It would be good to provide corroborating evidence regarding the heat

content decomposition, such as an independent calculation of total

heat content variability.

Abstract: Heat transport is stated to not always be responsible for

ocean heat content changes. Doesn't it have to be either the heat

transport or the air-sea flux, given thermodynamic energy

conservation? Heat content storage will be the residual of these terms. It is puzzling to consider where the thermodynamic

energy is transported. The manuscript would benefit with a closed

energy budget analysis, which may require the deep ocean and/or the

Arctic to accurately assess where the energy goes. 

Minor issues:

L29: does this sentence equate objective analysis and ocean reanalysis? They are normally considered to be distinct.

How are ocean heat content changes related to algorithmic changes between OFES 1 and 2?

L65: consider changing "numerical running" to "numerical simulation".

How did the authors justify restricting their analysis to the upper 1400 meters? Their rationale following Emery (2001) and Wunsch (2011) is not compelling. Did these previous works suggest that ocean disequilibrium occurs suddenly at 1400 meters depth?  (That would be surprising.) During the time period of interest, i.e., 1950-present, why should the deep ocean be in equilibrium? 

For comparison purposes, does the NCEP reanalysis give air-sea conditions every 6 hours, as opposed to 3 hours in OFES 2?

L86: "Validation" is not possible with EN4 as it is also an incomplete and uncertain product.

Figure 1: What happens in the Arctic? What error is incurred by eliminating the Arctic?

Consider changing ZJ/yr into the more relevants quantity of W or watts per square meter of the Earth's surface.

Figure 2 is fascinating, if correct. What is going on with OFES2?

L161: dividing by 56 "years".

What does "SP" stand for?

L242: 10 to the 6th power

Does OFES 2 fit surface data (i.e., SST)? 

Table 4 doesn't seem very useful with the inexact metrics for the water-mass source properties.

This study provided a comparison of the simulations results by OFES1 and OFES2 against EN4 objective analysis product. There are many apparent differences between OFES1 and OFES2, between OFES and EN4. The authors tried to decompose the local temperature changes to the vertical displacement of neutral surfaces and the changes along the neutral surface. They further investigated the surface heat flux and heat transport to better understand the difference. Overall, I think most of the texts in this paper are devoted to present the differences rather than providing insights, therefore, the biggest issue is that the reasons behind all these differences are poorly known. Further, many of the analyses are not tenable in my opinion (see below). The conclusions of this study are not supported by the results. I could not suggest publication of the manuscript in the present form.

Major issues:

1. The decomposition into HV and SP. One major conclusion of this paper is “There was an OHC increase in most of the global ocean over a 57-year period, mainly a result of vertical displacements of neutral density surfaces.”. However, I don’t think it is a robust conclusion given the fact that neither OFES1 nor OFES2 well simulate the OHC changes globally or at each major ocean basin.
2. The investigation of heat flux and heat transport are not well designed and not useful. To examine the mechanisms for the change of OHC, you have to check the trends in heat flux and heat transport, not the climatology field (Figs. 10, 11). In another word, you have to know where more heats are input into the ocean and how they are transported.
3. The water mass analyses in section 3.3 are also problematic, because the water masses are defined by the density or the temperature/salinity range as in Tables 4,5, however, the figures 6 and 7 are presented at z-coordinate, so the discussions are very confusing and not corresponding to the plot.
4. Section 3.2. The zonal integrated OHC, HV and SP. This section superficially described the results without any in-depth analyses or insights. It is not useful for the audience.
5. Section 3.1. Why not also provide the global or basin time series for surface heat flux for comparison? Globally, the heat content change is balanced by surface heat exchange. The decomposing into HV and SP does not help to understand the mechanisms here.
6. I expect an answer of why is OFES2 so different from OFES1, so a formal ocean heat budget analysis should be done.
7. Another conclusion “However, these differences, more specifically in the heat transport, were only partially responsible for the OHC differences.” Is not tenable, because I did not see an analysis for ocean heat budget, and the current analyses are wrong because only climatological heat flux and transport are shown.
8. Final conclusion in the abstract “The marked OHC differences may arise from the different vertical mixing schemes and may impact the largescale pressure field, and thus the geostrophic current”. This is a full speculation without any evidence.
9. Is the decomposing of EN4 data into HV and SP consistent with previous results? How large is the uncertainty behind the decomposing method given the data errors?
10. For zonal integrated OHC analyses, why some regions models are closer to observations and some places are not? What are the possible reasons and what are the implications? Again, an ocean heat budget analysis at each zonal band might help to identify if the difference comes from surface or ocean heat transport.
11. Section 3.4: diving the ocean by 0-500m and 500-1400m will cross-cut several different water masses. It is really strange to use vertical levels in water mass analyses.

In summary, the manuscript looks like a superficial analysis of model results and I don’t think it will help the audience if the paper is in present form.