the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 09 Feb 2024
Present-day correlations insufficient to constrain cloud albedo change by anthropogenic aerosols in E3SM v2
Abstract. Cloud albedo susceptibility to droplet number perturbation remains a source of uncertainty in understanding aerosol– cloud interactions, and thus climate states both past and present. Through E3SM v2 experiments, we probe the effects of competing parameterized processes on cloud albedo susceptibility of low-lying marine stratocumulus in the Northeast Pacific. In present-day conditions, we find that increasing precipitation suppression by aerosols increases cloud albedo susceptibility, whereas increasing cloud sedimentation decreases it. By constructing a hypothetical model configuration exhibiting negative susceptibility under all conditions, we conclude that cloud albedo change due to aerosol perturbation cannot be constrained by present-day co-variabilities in E3SM v2. As such, our null result herein challenges the assumption that present-day climate observations are sufficient to constrain past states, at least in the context of cloud albedo changes to aerosol perturbation.
- Preprint
(2064 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-366', Anonymous Referee #1, 05 Mar 2024
Peer review for Atmospheric Chemistry and Physics
Title: Present-day correlations insufficient to constrain cloud albedo change by anthropogenic aerosols in E3SM v2
Authors: Naser Mahfouz, Johannes Mülmenstädt, and Susannah Burrows
In this manuscript, the cloud albedo susceptibility to cloud droplet number concentration (CDNC) is analyzed in an Earth system model (E3SM). Following the method of Zhang et. al. (2022), the susceptibility is analyzed in LWP-CDNC state space where the brightening and darkening regimes can be attributed to the cloud adjustments that are likely in those regimes given the cloud state. Starting from a default simulation, using a case based in the NE Pacific, the regimes are not so clear for the model result here as compared with the satellite composites (Zhang et. al., 2022). The manuscript then investigates two cloud adjustments, precipitation suppression and the sedimentation-entrainment feedback, through process denial and then process scaling. In both cases, scaling the process up increases the intensity of the response to aerosol perturbation. One of these experiments, where the dependence of autoconversion on droplet number is removed and fall speeds are 4x higher, is then used to compare pre-industrial aerosol conditions with present day. In both cases the cloud albedo susceptibility is negative throughout the whole LWP-CDNC state space however the increase in CDNC from PI to PD still results in an increase in albedo and hence the key result.
This manuscript contains an interesting set of experiments that nicely build on the work of Zhang et. al. (2022) and Zhang and Feingold (2023), and it is written to a high standard. The figures are clearly presented and easy to understand given the captions and analysis explanation. However, although I appreciate the brevity of the manuscript, it comes at the expense of clarity in several places. As the reader, I had to work quite hard at times to follow the logic, which does appear sound but could benefit from helping the reader fill in the gaps. My main comment on the method itself is about the presentation of this as a constraint method, which I think requires some clarification on how the cloud albedo change would be constrained. My recommendation to the editor is that this manuscript is accepted subject to minor revisions surrounding the presentation of the work.
Minor comments
1) The main finding of this manuscript is that the present-day correlations are insufficient to constrain pre-industrial albedo change due to aerosol perturbations, but as a reader I am struggling to understand this key point. The result shows that using the case of enhanced sedimentation-entrainment feedback produces a negative cloud albedo susceptibility in both PI and PD and as such one would expect that the increase in aerosol from PI to PD would decrease cloud albedo, which does not occur. This is shown and described clearly. However, supposing cloud albedo had done as expected, how would the result be used to constrain cloud albedo change due to aerosol perturbations? My understanding of “constrain” is reducing the uncertain range of some value through ruling out implausible regions. Here it seems to be used more to do with whether the model is representative of what we would expect. The finding that the model does not respond according to this susceptibility is interesting and significant, but it seems to be more about a systemic error.
2) The sedimentation experiment is one of the points that I think would benefit from more explanation. For the precipitation suppression, what is expected to happen and why is clear. But for the sedimentation the reader could be assisted in understanding the expectation. The paragraph beginning at line 137 describes what is done and the effect seen and the paragraph beginning at line 144 seems like it is going to explain why that might be so. But it only says that increasing the fall speed is expected to increase the sedimentation-entrainment feedback, which has already been shown in the result. As the reader, I have had to sit and think about what I would expect to happen physically.
I expect that in denying sedimentation, or lessening the fall speed, the droplets hang around longer in the cloud top region and therefore would cause a stronger darkening since they have more time to evaporate and cause the entrainment feedback. In fact, it seems to me that there is increased darkening in Figure 3, leftmost plot, but this is not mentioned in the discussion. Following the logic for lessening the fall speed, I would have expected that increasing the fall speed would decrease the sedimentation-entrainment feedback since the droplets are removed from the cloud top region before significant evaporation can take place. I can see the result in the rightmost plot shows the opposite, in line with what the authors postulate. The authors could consider taking the reader with them on why they postulate this, even if definitively proving the mechanism is outside the scope of this study. Are there any readily available diagnostics that could help explain this?
3) Following on from the above, the brightening in the sedimentation denial experiment is presumably a strengthening of the Twomey effect. The authors might consider highlighting that this is most likely the Twomey effect and again suggesting why we see an increase in this effect related to the suppression of sedimentation. Perhaps because the smaller droplets are remaining at cloud top?
4) In examining the precipitation suppression result, line 129-130 reads:
“It is evident that precipitation suppression denial (leftmost panel) increases the prevalence and magnitude of cloud darkening due to increasing droplet number. Moreover, the precipitation suppression experiments show a gradual increase of cloud brightening as precipitation suppression is strengthened from left to right, culminating in mostly brightening clouds in the rightmost panel.”
I have spent some time looking at Figure 2 and I am struggling to see any difference between the first three plots. The authors could consider quantifying the brightening/darkening in some way to justify what is described in the text.
5) The introduction is well written and very concise, however as the reader I want a little bit more to motivate the study, the approach taken and to fill in some blanks.
i) The authors could consider adding a fuller description of the sedimentation-entrainment feedback in paragraph 2. Not all readers will know what this is, and since the rest of the paragraph has explained the other adjustments clearly it is a shame to leave this one as a vague point. I notice that it is described later when analyzing the results, but that could even be moved up to the introduction and then referred to later.
ii) Also in paragraph 2, the adjustments start with “For thin, non-precipitating clouds” but does not go on to point out where thicker or precipitating clouds come into it. Perhaps the authors could consider adding something along the lines of “for thicker clouds likely to precipitate” somewhere in the lifetime effect.
iii) A description of the outcomes from Zhang et. al. (2022) and Zhang and Feingold (2023) does not appear until the first paragraph of the results section, but again, the authors could consider having an overview of their findings in the introduction. Especially because the paragraph beginning on line 25 talks about the different regimes and patterns they find, as the reader I feel left in the dark about what these are. This would follow on really nicely from the previous paragraph where these physical adjustments are described.
iv) The authors could consider adding some more references to the introduction, particularly in the first paragraph when discussing cloud adjustments (second sentence). They could also consider adding an extra sentence stating the current understanding of the cloud radiative forcing in that overall, it is negative.
6) The methods section is quite comprehensive with an extensive description of the setup of the simulations. The authors could consider including an equation to show how the fall speed is used in the sedimentation parameterization. The autoconversion is partly given, but they could also consider showing the full equation, including the dependence on specific humidity since it is a fairly simple equation.
References
Zhang, J. and Feingold, G.: Distinct regional meteorological influences on low-cloud albedo susceptibility over global marine stratocumulus regions, Atmospheric Chemistry and Physics, 23, 1073–1090, https://doi.org/10.5194/acp-23-1073-2023, 2023
Zhang, J., Zhou, X., Goren, T., and Feingold, G.: Albedo susceptibility of northeastern Pacific stratocumulus: the role of covarying meteorological conditions, Atmospheric Chemistry and Physics, 22, 861–880, https://doi.org/10.5194/acp-22-861-2022, 2022
Citation: https://doi.org/10.5194/egusphere-2024-366-RC1 -
AC1: 'Reply on RC1', Naser Mahfouz, 07 Mar 2024
We thank the Referee for the thorough and insightful review! We will revise the manuscript accordingly very soon.
In the meantime, I wanted to address one important issue right away. In Item 4, the Referee rightly points out that our description of Figure 2 wasn't evident in the figure itself. That is because I made an error and accidentally repeated the first three panels. I apologize to the Referee for spending undue time reviewing this specific figure. I am attaching the correct figure to this quick reply, juxtaposed with the wrong figure in the current manuscript. The mistake took place when I was addressing a coauthor's recommendation to increase the font on the figures, and I mistakenly repeated the same (first) panel three times. The description of the figure as well as the remainder of the text are unaffected. I also double-checked all other figures in the manuscript, and they are not affected. Again, I am sorry about this mistake.
-
AC1: 'Reply on RC1', Naser Mahfouz, 07 Mar 2024
-
RC2: 'Comment on egusphere-2024-366', Anonymous Referee #2, 08 Mar 2024
-
AC2: 'Reply on RC2', Naser Mahfouz, 14 Mar 2024
We thank the Referee for the thorough and insightful review! We will revise the manuscript accordingly soon. We will post our responses and revised manuscript upon the closing of the discussion period (after 22 Mar 2024).
Citation: https://doi.org/10.5194/egusphere-2024-366-AC2
-
AC2: 'Reply on RC2', Naser Mahfouz, 14 Mar 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 332 | 100 | 17 | 449 | 9 | 10 |
- HTML: 332
- PDF: 100
- XML: 17
- Total: 449
- BibTeX: 9
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1