It is an interesting work to integrate mercury cycling in the CESM. We recommend minor revision after addressing the following comments.

Abstract, “but they have limited capacity in predicting the future”

It is better to simply explain the reasons why most global atmospheric mercury models have such limitations.

“One advantage of our online model is that the concentrations of Hg oxidants are calculated online.”

What is the time resolution of the online Hg oxidants concentrations?

“Aerosols in CAM6-Chem are represented using the four-mode version of the Modal Aerosol Model (MAM4), including sulfate, black carbon, primary organic matter, secondary organic aerosols, sea salt, and mineral dust (Liu et al., 2016). Secondary organic aerosols are treated

using a volatility basis set (VBS) scheme, which is described in detail by Tilmes et al. (2019).”

Aerosols concentrations significantly impact the species transformation of Hg in atmosphere? How are the performances of MAM4 and VBS in predicting the concentrations of aerosols as well as secondary organic aerosols? 

“The natural emissions are derived from the average of a 5-year simulation in GEOS-Chem, including geogenic, biomass burning, soil, snow, and vegetation emissions.”

Earth system has a natural advantage to couple mercury transfers between different spheres. It is better to calculate the natural emissions in CESM instead of using the results derived from GEOS-CHEM directly.

“The best match with the available observations can be obtained by adjusting the photoreduction rate coefficient” 

Please compare the adjusted photoreduction rate coefficients with observations to verify the reasonability of the adjusted results.

“The representation of the main oxidants (e.g., O3 and OH) of Hg0 have been greatly improved and are more comprehensive in the CAM6-Chem, comparing with its predecessors CESM or CCSM (Lamarque et al., 2012; Emmons et al., 2020).” 

Briefly describe the results please.

“The global total Hg emissions from all sources are about 7000 Mg a-1, two-thirds of these are natural or legacy emissions. In this study, the rapid re-emission of deposited Hg0 is included in the land emission and the Hg0 dry deposition to the ocean is replaced with the net Hg0 evasion.”

Please introduce how legacy emissions and re-emissions are considered in the method part.

Figure 1, “HOHgI and Hg(OH)2 formed by the oxidation of Hg0 by OH are relatively stable” cannot persuade the readers because the stability of oxidants does not necessarily relate to the significance of pathways. For example, HgBr· is not stable in atmosphere but it is a significantly mid product in the two-step Br oxidation process and Br· is argued as important oxidants. The oxidation pathways are significantly different from previous studies (e.g., Horowitz et al., 2017). Please compare with previous studies and explain the reasons as well as the reasonability of the results. In addition, does the dominant oxidation pathways differ across regions? Please briefly introduce the results?

Figure 4 Please present the uncertainty range of model results.

Review of “Earth system modeling of mercury using CESM2: part 1. Atmospheric model CAM6-Chem/Hg v1.0” by Zhang and Zhang (Paper #GMD-2021-380).

In this study, the authors used the CAM6-Chem atmospheric component in the CESM model to implement the mercury chemistry that includes oxidation of Hg⁰ by both Br and O₃/OH with reaction rates taken from Horowitz et al., (2017). They calculated the global atmospheric Hg budget. The model was evaluated against observations for total gaseous mercury (TGM) and deposition (dry and wet) of Hg. They further discuss the latitudinal and seasonal variations and the contribution of different processes to the variations in the two hemispheres. The paper can be accepted after addressing the following comments:

General comments:

1. Shah et al., (2021) had a major revision to the GEOS-Chem mechanism described in Horowitz et al., (2017), how do the authors account for those changes? It is relevant to include a discussion on the mercury chemistry implemented in this study with Shah et al., 2021, which is the most updated mercury chemistry in a global model.
2. Why were not the emissions from Streets et al., (2015) used in the study? The Global Mercury Assessment 2018 used it in their report and in Shah et al (2021).
3. Which version of the GEOS-Chem model output was used for the natural emissions? Add the details in the manuscript.
4. Is there a specific reason for choosing the simulation period 2011-2013?