This manuscript provides in a compact way a new module to describe secondary organic aerosol concentrations, and its impact on radiation in a simple, and computationally efficient way, using the complex parameterization as in CAM-chem as a reference. This replaces a previous simplistic implementation which has known limitations, with only a marginal increase in computational costs. Furthermore, it allows a more closer alignment with the full-chemistry parameterization in future.

The authors present in a clear way the improvement that is gained compared to the original module with respect to the reference (full chemistry) configuration from a climatological perspective, i.e. the main application area of this new module. 

While reading the manuscript, I wondered why a single 1-day e-folding lifetime for the SOAE is used, (line 166) as I believe it would make much more sense (without a significant increase in computational cost) to make this loss rate dependant on the seasonal (and diurnal) change in OH. This could lead to some changes in the seasonal cycle of the SOA burden, and its profile, I can imagine. But the authors have also noticed this as a point of future improvement. 

This manuscript is fit for publication in GMD, if possible after consideration of the following comment. Table 2: Can the authors report on the changes in SOA lifetime (or SOA production) across the different configurations? If this does not fit in the table, this could also be reported as part of the text for the 2013-experiments only.

Technical comments:

Figure 1, CAM-chem description: Here SOAG0 is in balance with SOA1, shouldn’t this be ‘SOA0’  (and likewise SOAG1 and SOAG2)? 

line 143: “after model tuning involving the aerosol indirect effect”

line 260: ”although the two CAM cases”

Jo and coauthors present an improvement to the CAM6 model that updates the treatment of SOA so that it is more consistent with the predictions of CAM-Chem. The team has chosen their updates judiciously and demonstrated impressive consistency in results for pollutant concentrations and radiative forcing compared with the CAM-Chem simulations. Important divergences like SOA at the top of the atmosphere seem to be almost completely resolved. The manuscript is organized well, and the figures capably illustrate the main points of the discussion. This paper should be published in GMD. I have a number of concerns I would like the authors to address first though.

1) The main criticism is that the discussion of results speculates often about the aspect of the update that is responsible for a particular improvement in results. For example, lines 271-278 attribute the improvement in vertical profile to the incorporation of the intermediate tracer and introduction of a time delay in SOA formation. But part of this could also be due to the change in deposition or enthalpy of vaporization. Because there are a limited number of parameters that describe the new SOA configuration, it would be instructive to implement them one at a time and present the influence each has incrementally on the final results.

2) The abstract lacks any quantitative information documenting the improvement in predictive power of CAM with the new SOA scheme. Please add some summarizing statistics.

3) Can the authors please add more information about the deposition scheme and parameters used for the loss of aerosol- and, particularly, gas-phase SOA species? Are Henry's Law values parameterized and used? Is pH considered?

4) What is the sensitivity of results to changes in the assumed saturation concentration and enthalpy of vaporization?

5) Section 2.4: Why are BC and POA particles emitted in the accumulation mode instead of the Aitken mode where most particles are emitted by anthropogenic combustion sources? What is the sensitivity to this assumption?

6) Section 2.4: If SVOCs and IVOCs are added to the model, can the authors explain how they know they aren’t double-counting carbon emissions with POA?

7) Section 2.4: It is surprising to me that the differences in Figs. 2b and 2c are due entirely to SOA aging. Can the authors be a little more specific about the mechanism of the change in deposition for POA and BC? Is this an impact of particle size change or change in bulk particle hydrophilicity?

8) Please add details to the methods description explaining what MAM4 is using for aerosol moments (e.g. 2-moment or 3-moment). Is the standard deviation of each mode a variable parameter? Also, does Aitken mode mass grow into the Accumulation mode?

9) Lines 154-162: More details should be given about how the yields were parameterized for the new SOA scheme. It is clear that the yields for the c* = 100 ug m^-3 bin should be reduced, but how did the authors choose 20% and what is the sensitivity of results to this parameter? Early in this paragraph, it is claimed that the yields for the four lowest bins are merely added up, but later the authors write that the yields are based on CAM-Chem results. Which of these are true, and what is the basis for the choice?

10) How sensitive are these results to the choice of aerosol activation scheme and cloud microphysics module?

Minor Comments/Typos:

1) Lines 43-45: It seems odd to introduce the VBS without citing any of the formative VBS literature references (e.g. Donahue et al., 2006, 2011, 2012; Jimenez et al., 2009, etc.)

2) Line 63: ‘the same’ --> ‘similar’

3) Line 112: ‘simple’ --> ‘first-order’ or ‘bimolecular’

4) Line 150: Although it is cited in Shrivastava et al. (2022), I recommend adding a specific reference to Lim and Ziemann (2009) since they showed this definitively.

References

Donahue, Neil M., et al. "Coupled partitioning, dilution, and chemical aging of semivolatile organics." Environmental science & technology 40.8 (2006): 2635-2643.

Donahue, Neil M., et al. "A two-dimensional volatility basis set: 1. organic-aerosol mixing thermodynamics." Atmospheric Chemistry and Physics 11.7 (2011): 3303-3318.

Donahue, Neil M., et al. "A two-dimensional volatility basis set–Part 2: Diagnostics of organic-aerosol evolution." Atmospheric Chemistry and Physics 12.2 (2012): 615-634.

Jimenez, Jose L., et al. "Evolution of organic aerosols in the atmosphere." science 326.5959 (2009): 1525-1529.

Lim, Yong B., and Paul J. Ziemann. "Effects of molecular structure on aerosol yields from OH radical-initiated reactions of linear, branched, and cyclic alkanes in the presence of NO x." Environmental science & technology 43.7 (2009): 2328-2334.

This paper presents a new simplified secondary organic aerosol scheme in the atmospheric component of CESM2 model (CAM6.3). The new approach aims to replace the previous simplified SOA formation scheme available in CAM with known biases compared with the more complex parameterization based on the volatility basis set (VBS) approach (CAM-chem). The authors describe the new scheme and the design strategies adopted to achieve closer results to CAM-Chem at a much lower computational cost. The paper is well structured and the figures and tables provide proper support to the description of the work. However, some additional details could be provided for a comprehensive description of the new scheme presented in the manuscript. This paper should be accepted for publication in Geoscientific Model Development after addressing the following minor comments.

Specific comments:

• The authors could provide some quantitative results in the abstract instead of using expressions like “the overestimation ... is greatly reduced” or “ the high bias ... is significantly reduced”.
• Provide some key parameters used in the new scheme for the sake of completeness in Table 1: (1) the Henry’s law constants of SOA gas-phase precursors used in the new scheme (is the ones described in Hodzic et al. (2016) Table 2 for the third volatility bin?), (2) the wet scavenging efficiency of SOA (set to 80%?), (3) the photolysis rate coefficient (set to 0.04% of NO2 one?), and (4) the uptake coefficient of the heterogeneous reaction with ozone (set to 10e-5).
• Line 114: Hodzic et al. (2016) mention that 20% of total NMVOC emissions (not including SVOC emissions) are assumed to be IVOC emissions. Could the authors elaborate more on how SVOC emissions are excluded from total NMVOC emissions and how double counting is avoided when using S/IVOC emissions derived from NMVOC and POA?
• Line 160: Could the authors clarify if the SOA yields from CAM-chem used to derive the values reported in the text for the new scheme are derived from Hodzic et al. (2016) Table 2?
• Only low NOx yields are used in the design of the new scheme. Some discussion about how the scheme could be extended to cope with high NOx regimes would be interesting, at least as a perspective for future works.
• Apart from the SOA yields, the authors should provide the resulting scaling factors applied to SOA precursor emissions. Figure 1 or Table 1 could be improved by detailing the key parameters of the scheme (e.g. emission scaling factors, H*, photolysis rate, ...).
• Line 184: In Section 2.4, the authors could provide some additional details about the linkages between BC, POA and SOA. The aerosols are assumed to be internally mixed in MAM4, is the total mass of one mode used in the gas-aerosol partitioning of SOA or only the total OA mass? How is SOA affecting the microphysical ageing of the mode? The reader would appreciate a brief description instead of simply providing a reference.
• Line 223: Additional budget metrics would be appreciated in Table 2 (i.e., production, deposition, lifetime).

Technical comments:

• Figure 1: soa3 is repeated in CAM-chem scheme
• Line 117: CLM is used in different places in the manuscript, consider defining the acronym like “the Community Land Model (CLM) version 5” here. The same applies for other acronyms like CAM or MEGAN.
• Line 225: missing capitalizing letter “Gas-phase SOA (SOAG) is substantially...”

This manuscript gives a generally clear description of the development of a simplified SOA scheme which is designed to mimic as far as possible the results of the more complex scheme (CAM-Chem). I think that the manuscript is worthy of publication after addressing the comments raised by referees. At this stage three referees have already delivered comments, so I will try to avoid making the same points. However, I do have some additional remarks:

1) The manuscript would benefit from version numbers or better labels for the original and new schemes. Terms like "NEW" (or "current simplified SOA") tend to age poorly, and somebody reading this manuscript in 10 years time might find that NEW was now OLD, and current wasn't current.

2) I miss discussion of these schemes compared to the real world, and indeed the text often makes it sound as though matching the more complex scheme is the same as improving model performance. For example, statements such as  "SOA underprediction during the ... winter time", or "simulates too much SOA globally" sounds as though we are comparing with reality, but the comparison is only with the complex scheme. Be explicit with such comparisons, and please add some comparisons with the real world!

p2, L20, and elsewhere. It is surprising that a new SOA scheme would affect BC and POA, so I would add a short explanation of why this happens here and in the main text.

p4, L43. The VBS is not the most detailed version of SOA mechanisms; indeed it is rather simple compared to e.g. Xavier e al (2019)

p4 L41. I would add that parameterizations are used because there are also many uncertainties in our basic knowledge of SOA formation. Also, why do you say "generally use"? All models use parameterizations!

p7, 2nd paragraph. It isn't clear here if "SOA" means both gas and particle phase, or just particle phase. You describe the gas-phase SOAG as as an "intermediate precursor of SOA" (L111), which makes it sound as though SOAG is lost to particulate SOA (SOAP?) with no return. In the VBS the SOAG and SOAP phase should be in equilibrium, so SOAG is no more an intermediate than SOAP. Throughout the manuscript it is not clear what the term "SOA" means. Please clarify and tighten up the notation.

p7, L114. Do these SVOC and IVOC assumptions add mass to the emissions, or are they just fractions of the original emissions? If 20% of NMVOC are treated as S/IVOC, are they effectively removed from the gas-phase (and ozone producing) mechanisms?

p8, 145. The new scheme isn't really explained here. Please refer to Fig. 1 (which is quite clear, except whether SOA means SOAP), and that the new scheme uses a 1-bin VBS.

p8, L135. Is dry and wet deposition not considered for particle-phase SVOC? Fig. 1 suggests it is.

p8, L150. Add "generally" before yield; many factors other than simple carbon number decide the yields.

p9, I found this first paragraph rather confusing. This text discusses VBS yields as though they are fixed quantities, but do you mean the yield of gas+particle phase compounds (SOAG+SOAP), or the yield of particle phase (SOAP) only? The particle yields depend heavily on the ambient absorbing mass, as well as temperature.

p9, L174. Why isn't deposition of particle phase SOA considered?  This statement contrasts with Fig.1.

p18, section 4. The conclusions should give a quick summary of what the "new SOA" scheme entails, e.g. 1-bin etc.

p19, L348. "improved the performance of .." = another example of confusing text for me, see point (2).

p20. Code and data availability.  As already noted by the editor, this manuscript should be associated with a doi where readers can access the code. And which model results are available at DASH. And what is DASH, and where is it? This was very vague!

p21-24, References. Many lack doi information.