The authors set up 2 modal aerosol models, each with both simple and full chemistry, in the Dept. of Energy’s Earth system model and compare the model output for 1991-1993. The object is to characterize the development of stratospheric aerosol from the Pinatubo eruptions from the 4 different aerosol models. The models are compared to AVHRR and HIRS observations.

The paper is mostly well written with a specific purpose but there are difficulties. The first difficulty is the limited comparisons with observations, just two sets of observations and two figures. Both figures raise questions about the comparison, but little discussion is given to the disagreements. A lot of papers have been published describing the post Pinatubo aerosol from a variety of instruments. Why aren’t additional comparisons made with a much wider set of data? The model has pretty fine resolution, so it should not be limited to comparisons to measurements with global coverage. The authors spend a lot of energy comparing the various size distribution modes from the different model configurations, but make no attempt to compare any of these size distributions to observed size distributions. The authors offer no explanation for their limited comparison with observations.

The second difficulty has to do with the figures, their discussions, and the language used to describe processes related to aerosol moving between modes. Details related to these issues follow here by line number. Within this are also a few minor suggestions.

119-120 In terms of size order aren’t the modes: Aitken, accumulation, coarse, rather than accumulation first? If so then they should be listed in that order.

Fig. 1 What happens to particles in MAM4 when they exceed Dg_hi (0.48 µm), which is not that large for sulfate particles following Pinatubo? See e.g. Deshler et al., GRL, 1992.

178-179 The parenthetical clause is so long the reader has lost the thread as to what limits the aerosol formation rates.

Fig. 2 and its discussion. What is the explanation for the aerosol in the Southern Hemisphere, which appears at most longitudes almost simultaneous with the Pinatubo eruption, particularly in MAM5FC? The presence of this aerosol clearly above background should be acknowledged and if possible explained.

284-286 Why does the smaller geometric standard deviation in MAM5 lead to more persistence? Is it because the particles are smaller in MAM5 compared to MAM4 and therefore less sedimentation? In any case there should be a sentence to describe the physics involved.

Fig. 4 and its discussion. The discussion mostly consists of describing the figure providing specific dates and sulfate burden peaks for the various modes. While perhaps these details are interesting they are all available in the figure for the interested reader. More interesting for the reader would be more discussion of the model / measurement discrepancies. Why do all models except MAM4FC over estimate the peak observed sulfur burden by 20%? Are the HIRS data reliable at the peak or are they suffering a saturation problem? Why does HIRS fall off so much faster than the MAM5 models in 1993? Additional interesting detail would be the range of median sizes involved in the large particle mode.

Because MAM4 doesn’t have a coarse mode and MAM5 has a very narrow accumulation mode, both Figs 4 and 5 show the same thing, no contribution in the accumulation mode (or very little) from MAM5 and no contribution in the coarse mode from MAM4. Why then show the accumulation mode at all? Combine the MAM4 accumulation mode and MAM5 coarse mode into one figure. MAM5 accumulation mode could be included as a dotted line in the coarse mode plot. Then the two models can be more easily compared in terms of sulfate burden carried in the large particle mode.

Fig. 5 Same comments as Fig. 4 but in addition the ordinates should all be the same scale (0-4 Tg S), so the relative contributions from the different latitude zones can be seen directly. Without that the reader immediately wonders about the Southern Hemisphere signal which seems to persist at high levels. The label on the ordinate is wrong. It should be Sulfur burden (Tg S). Label the rows in some other way or describe them in the figure caption. Again combine accumulation and coarse mode into one plot then there will only be three rows. The latitudes of the eruptions should be listed in the figure caption.

331-332 What is the reason for this oscillation, when the sulfur burden is declining everywhere else? A somewhat similar oscillation, but offset, is observed in the Northern Hemisphere. Why does the sulfur burden persist to 1994 in the Southern Hemisphere while it decays everywhere else?

341-343 This point would be a lot clearer for the reader if the ordinate scale on all plots was the same. But this statement, “Starting from 1993, the stratospheric burden south of 30 S in MAM5 begins to make up at least half of the total between 80 S and 80 N (about 0.8 Tg S out of the total of 1.7 Tg S)”, doesn’t make sense. Between 80 S and 80 N includes 30 S – 30 N, so the total sulfur burden in early 1993 for MAM5FC is ~4.1 Tg S.

349-350 Suggest rephrasing to. “The tendencies are the integrals over three-dimensions of: all longitudes, …”

Fig. 6 is problematic. There are too many rows making the figure so small that most readers have to blow it up to see it. This could be fixed by breaking it into two figures: the first containing the first 4 rows, the next containing the last 3. Another suggestion is to combine the right and left panels into a single plot with separate ordinates on the right and left. Since the left shows a rate and right shows an accumulation the lines will generally not overlap but rather complement each other. Again the ordinate labels are incorrect. They should be tendency (kg/s) for the left plots and their integrals (or cumulative) (Tg S) for the right. Include the name of the row as a label in each plot. The labels RNMxx are too tied to the inner workings of the model, “renaming”. But physically what is happening? The particles are growing to the next largest size distribution mode. If labels were added to the plots to identify them RNMaa could become Aitken->Accumulation mode and RNMasc Accumulation->Coarse mode.

Fig 6 e, f) shows the growth from Aitken to accumulation mode presumably by several processes including growth by condensation and coagulation, correct? If that is the case then why are the ordinate scales on COAG so much larger than for RNMaa? Both are showing mass loss rate and total mass lost. It seems it should be the other way around with COAG less than RNMaa. Why is this one process coagulation singled out for a special plot?

Fig 6 i – n) Condensation? Why are these processes now called condensation? Condensing from what? Weren’t these earlier called renaming, which is also not that helpful or descriptive. Isn’t this particle growth from one mode to another? Ordinate problems again. What is on the left and right ordinates? Is it again rate (kg/s) and cumulative mass (Tg S)? The reader doesn’t know and the figure caption does not help.

351-415 Again the figure discussion consists primarily of describing the figure, pointing out maximum values and dates when they occur. Relatively little is describing what can be learned from the figure which is the importance of the figure. Here and elsewhere, if these dates and amounts are particularly important organize them into a table. Then they could really be compared. It is not clear how listing them in the text helps the reader.

Figs. 4-8, 11 The plotting for these figures could be made much more intuitive, so the reader doesn’t constantly have to refer to the legend to remember which is which. It would be quite easy to do. Use one color for MAM4 and one for MAM5, then one line style (e.g. solid) for FC and another (dashed or dotted) for SC. Then each figure can be immediately understood without referring to the legend but once.

Figs 7 and 8 suffer from the same problem as Fig. 6. There are too many panels and they are too small. What are Figs. 7 and 8 adding to what we learned from Fig. 6? Are all these rows necessary? Which ones are the most informative?

420-430 In fact the discussion of Figs. 7 and 8 acknowledges that not much new is added. “The same patterns as above apply between 30 S and 30 N. Above 30 N and below 30 S the same signal from the Pinatubo is still present, though slightly delayed due to the time that it took for the aerosol to transport poleward. Signals from other eruptions (Spurr and Lascar) are also present” Then a few interesting differences are discussed. Just show the interesting panels and combine Figs 7 and 8 into one figure with the few interesting panels.

503 condensation? Same questions as above. What does this mean?

503-505 What is the reason for quoting these numbers? How will the reader use such information rather than the already stated comparison about the difference rates? Too detailed.

506 Here COAG is separated from RNMaa, but aren’t both processes doing the same thing? There is only the transition from Aitken to accumulation. Is it important how it happens? If so why isn’t that mentioned earlier?

509 deposition? Does this mean sedimentation out of the stratosphere? Generally deposition refers to losing aerosol due to contact with a surface.

Overview:

This paper explores simulations of the Pinatubo eruption using different configurations of the Modal Aerosol Module (MAM) in E3SMv2. MAM is used with 4 modes and extended with a 5^th mode to better capture large stratospheric aerosol particles. Additionally, both simplified and full chemistry modes are tested.  The paper is well written and structured with clear explanations of the differences for full chemistry due the importance of OH reactions and inclusion of a larger mode to more accurately capture growth and sedimentation processes. I think this paper is a good contribution to the literature on prognostic aerosol modelling in the stratosphere and I would recommend publication with minor clarifications provided it falls within the scope of ACP.

General Comment:

I think the article is well written and a nice advancement to stratospheric aerosol modeling but may be better suited in its current form for a journal such as Geoscientific Model Development. While the differences in aerosol loading due to model configuration are made clear, it is left to the reader to interpret how this improves understanding of atmospheric chemistry or physics. The authors provide in-depth discussion on the relative importance of coagulation, nucleation and renaming/growth in the model, but have little discussion on the physical processes this may help resolve. Similarly, few comparisons are made with measurements for AOD, particle size or radiative flux, with no discussion given to possible sources of disagreement, or what implications these results may have for observations (e.g. the assumptions going into the HIRS results used here). Personally, I think addressing any of these points would help expand the applicability of the paper to a more general audience. 

Specific Comments:

Line 50-65: This geoengineering section seems a bit out of place to me. I'm sure this work has implications for geoengineering studies, but no indication of exactly what those may be is provided. I recommend clarifying the link to this work or removing this paragraph. Perhaps the geoengineering discussion and the link to this work would be better placed in the discussion/conclusion?

Line 415: “MAM4 also generally has stronger nucleation and coagulation processes than MAM5.” From Figure 6 it isn’t clear whether MAM4FC has greater coagulation tendencies for physical reasons or if it is just due to the increased NUCL. This is discussed in the conclusion of the paper (Line 503), but I think should be mentioned here. I would suggest rewording to something like: “MAM4 also generally has stronger nucleation than MAM5, and due to these higher concentrations, increased coagulation processes as well” Or, if there are other reasons for increased COAG then this should also be discussed.

Line 490: “large differences in both the temporal variations and the spatial distributions of sulfate concentrations” It is difficult to tell from Figures 7/8, but in Figures 9/10 there doesn’t appear to be much change in spatial distribution. Both MAM4 and MAM5 show large increase in the tropics and later transport to the NH. Some expansion on the spatial differences the authors are referring to would be welcome.

Line 521-522: If the use of full chemistry and MAM5 helped improve agreement with AVHRR, why is the TOA flux more comparable to Brown (2024) and Mills (2017) results than the MAM4 version? Is this related to the geometric standard deviation that was used?