the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Oct 2025
Volcanic Aerosol Modification of the Stratospheric Circulation in E3SMv2 Part II: Brewer–Dobson Circulation
Abstract. Great attention has been paid to the short-term climate response following large volcanic eruptions, in order to understand effects on zonal winds, the polar vortex, and surface temperature across latitude. In contrast, several works have shown that evidence of volcanic forcing can persist for much longer in the stratosphere's chemical composition, even after the instigating aerosol population has dissipated. Heating by volcanic aerosols accelerates tropical upwelling, and thus drives an acceleration of the Brewer–Dobson Circulation (BDC), and enhances troposphere--stratosphere mass exchange. Even after tropical motion returns to its climatological mean, the anomalous mass exchange remains detectable in the stratosphere for several years. In this work, we use an age-of-air (AoA) tracer to diagnose stratospheric composition changes following the simulated 1991 Mt. Pinatubo eruption. Specifically, we employ simulation ensembles from the E3SMv2 climate model to identify statistically significant effects on zonal-mean AoA. In addition, we use the Residual Circulation Transit Time (RCTT) diagnostic to separate the effects of advective transport and mixing. We find that the Mt. Pinatubo eruption lowers AoA in the middle-to-upper stratosphere globally, primarily due to an accelerated residual meridional circulation. We also observe a localized increase of AoA near 20–100 hPa in the hemisphere opposite the eruption, which we attribute to a dampening of the seasonal BDC cycle by the volcanic aerosols. We suggest that a dampened seasonal BDC cycle is perhaps a generic result of any heating process driven by aerosols that evolve on timescales beyond seasonal in the meridional plane.
- Preprint
(6067 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 12 Dec 2025)
- RC1: 'Comment on egusphere-2025-4598', Anonymous Referee #1, 26 Nov 2025 reply
-
RC2: 'Comment on egusphere-2025-4598', Anonymous Referee #2, 30 Nov 2025
reply
Review of Volcanic Aerosol Modification of the Stratospheric Circulation in E3SMv2 Part II: Brewer–Dobson Circulation by Hollowed et al.
General comments: This study uses the E3SMv2 climate model to investigate the impacts of sulfate aerosols from the 1991 Pinatubo eruption on the Brewer-Dobson Circulation. The manuscript is very well written and follows a clear logical structure. The setup of the simulations is well described, and so is the analysis framework. The results are well and thoroughly presented. I strongly recommend publication and only have several minor comments as listed below.
Specific comments (minor):
- line 1 and lines 16-17: Since volcanic eruptions are very diverse, I would appreciate if the authors would specify what kind of eruptions they are investigating upfront (for example, in the case of this study, explosive volcanic eruptions injecting large amount of SO2 into to stratosphere).
- line 12: I would recommend specifying that the hemisphere opposite the eruption in the southern hemisphere as the authors do not simulate SH eruptions. This would be consistent with the wording of point 3 in Section 7.
- line 66: The authors define the abbreviation SAI later in the manuscript but I would appreciate they did it here already.
- lines 89-90: I assume the model is fully coupled.
- line 98: Although the authors specify the altitude and the latitude of the volcanic SO2 injections in Part I, I would appreciate if they did it here as well.
- line 171 (Eq. 11): The authors explain the variable X as the forcing by ‘parameterized processes and diffusion’, i.e. unresolved forcing. Later (for example in Fig. 5c and lines 247-251) they discuss this parameter only in terms of diffusion. Does this mean that unresolved processes other than diffusion are negligible in the TEM framework? If so, I would be interested in reading about it. And if not, I would be interested in reading more about which processes are not resolved and their relative importances.
- line 220: ‘[...] whereas our reference point is close to the surface.’ Can you say that 700 hPa (ca. 3 km) is close to the surface? I suggest replacing ‘close’ with ‘closer’.
- line 254: For clarity, I would appreciate if the authors specified panel (a) of which figure they are referring to (Fig. 3a).
- line 257: For clarity, I would appreciate if the authors specified in which figure the ‘relatively small local diffusion tendencies’ can be seen (Fig. 5c).
- line 259: For clarity, I would appreciate if the authors referred to Fig. 5e in the context of the net tendency.
- lines 275-277: From Fig. 6, it looks like that not only has the significance of the features discussed ‘notably diminished by August of 1992’, but almost disappeared in the tropics (as the authors note in line 296).
- Figure 13: Since the bulk of study focuses on the Pinatubo eruption (10 Tg SO2), I wonder why the authors choose to normalise the relative max impacts in the lower sub-panels to the results from their 15 Tg SO2 simulations.
- lines 486-487: The authors expand on their study of the Pinatubo eruption by simulating similar eruptions (in terms of location and injection altitude) of different emission magnitudes (Section 6). In Section 7 (lines 486-487), they further suggest that a future study could explore the sensitivity of their results to variations in eruption latitude and season. I would like to add that exploring the sensitivity to injection altitude would be interesting. Particularly in the light of the recent study by Toohey et al. (https://doi.org/10.5194/acp-25-3821-2025) which showed how the lifetime of volcanic aerosols in the stratosphere is very sensitive to the injection altitude.
Citation: https://doi.org/10.5194/egusphere-2025-4598-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 154 | 34 | 19 | 207 | 11 | 11 |
- HTML: 154
- PDF: 34
- XML: 19
- Total: 207
- BibTeX: 11
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of Volcanic Aerosol Modification of the Stratospheric Circulation in E3SMv2 Part II: Brewer–Dobson Circulation
The article discusses the modification of the stratospheric Brewer-Dobson circulation in response to the Pinatubo volcanic eruption, building upon the published results (Part I) on the stratospheric dynamics response. The results show that the heating-induced enhancement of tropical upwelling (~15ºN) results in reduced mean age of air (AoA) throughout the middle and upper stratosphere, while in the lower stratosphere (below ~30 hPa) in the SH there is a localized increase in AoA. The fact that the eruption is located in the NH tropics and the seasonal cycle in turnaround latitudes are found to explain the increased aging in the SH. A comprehensive analysis of the causes for AoA changes is carried out by combining the local tendency perspecive using the TEM budget of AoA, as well as the integrated transport perspective using RCTT and aging by mixing. The paper is very well written, the relevant literature is largely discussed, the analyses are well described to allow for reproducibility, the figures have high quality and support the results, and the conclusions are supported by the analyses shown. I therefore recommend publication after minor revisions. I do have some minor comments, which are aimed to improve the clarity of the paper and I strongly recommend addressing them.
Minor comments
- The top of the model at 60 km is located in the mesosphere and low in comparison with other state-of-the-art stratosphere-resolving models. I suggest adding a comment on the possible effects of this (relatively) low top, for instance on the BDC driving by gravity waves, the sponge layer, …
It could be perhaps affecting the longest-lived air parcels and explaining the misbehavior mentioned in the paper (missing trajectories, bouncing between the two hemispheres, e.g. L202-204).
- L115-125 (and Fig. 1): This discussion on the simulations is confusing. Why are a 3-year and a 7-year simulations needed? Why not carry out a single control simulation spanning the entire time period 1981-1998, and branch from it the different ensemble members, with and without perturbation? Please explain this in a simple and concise way in the paper.
- Throughout the paper, the term ‘tracer transport’ is used to denote ‘advection by the residual circulation’, in opposition to mixing. In my view, both advection and mixing lead to ‘tracer transport’ (while mixing alone does not lead to mass transport). Since this term is commonly used in the literature, and is the standard nomenclature of the TEM terms in Eqs. (5) and (6), I recommend to use it instead. For example in L154, L247, L254, L322, Fig. 10 title of panel (b) and figure caption.
- Fig. 4: Why do some trajectories seem to stop in the tropical lower stratosphere (below ~30 hPa) with transit times below 1 year?
- L281-282: ‘while the deep branch in the summer hemisphere is weakened’. I suggest specifying ‘the upwelling in the summer hemisphere is weakened’.
- The enhanced residual circulation is seen only above ~10 hPa in boreal winter. At lower levels it is actually weakened, as shown by the negative streamfunction response over the region of positive control streamfunction in Fig. 6. In the summer months instead there is an acceleration also in the lower stratosphere.
- Figure 9: In panel (c), the RCTT in the control simulation trajectories seem to stay constant after crossing the 20 hPa level, is this because the colorscale gets saturated? In this case it could be extended (otherwise it seems that they arrive to the final point with the same RCTT as the perturbed ensemble).
- Figure 10: In Fig. 7e the SH older-age region (SHLS) extends to July 1993, so Fig. 10 should be extended until that date to show the ‘abrupt’ termination of this feature. Is this what is referred to in L364-366? In that line it says the termination happens in June 1992, which is not what we see in Fig. 7e. Please clarify when does this aging terminate and show it in a figure (also ‘terminates after one year’ in L373 should be changed to ‘two years’?).
- Fig. 10 (less important than previous comment): For consistency with Figure 7e, this figure could show 40 hPa instead of 30 hPa. Finally, the black arrows in this figure are not necessary (the upwelling/downwelling regions are identified by the w*=0 contours) and in my opinion make the figure confusing, so I suggest removing them.
- Figure 12: Suggested modification of the schematic: Remove the annual mean picture, does not help to make the argument and it is not a realistic situaltion. Mark the turnaround latitudes (with vertical dotted/dashed lines, or shading the upwelling region), since the important point here is whether the anomalous downwelling falls into the climatological upwelling or downwelling region.
- Section 5.2: The damping of the seasonal cycle of AoA is interesting but could be shown more clearly. Fig. 11 shows that the eruption impact on AoA has the same sign as the boreal summer climatological tendency. But how that implies a reduced seasonality is not clear. Could you maybe show the reduced seasonal cycle for some specific region where it is most evident?
Technical:
- L199-201: This is already discussed above, consider reducing or removing and referencing back.
- ‘dampening’ should be ‘damping’ in L12 and L34.
- L380: ‘impact of’ should be ‘impact on’