the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Mar 2024
Representing the effects of giant aerosol in droplet nucleation in E3SMv2
Abstract. Giant aerosol, i.e., those with diameters larger than 1 μm, can form large droplets via condensational growth to sizes similar to drizzle particles without being activated. In this study, we assess the impacts of giant aerosol on clouds, precipitation, and radiation when activated giant aerosol are directly categorized as raindrops using the U.S. Department of Energy’s Energy Exascale Earth System Model version 2 (E3SMv2). We find that categorizing activated giant aerosol as raindrops reduces cloud liquid water path by 11.38 % globally, with most pronounced reduction in the mid-latitudes. We also find that this approach improves model's ability to simulate the positive correlation between surface rain rate and coarse mode aerosol concentration in regions of low precipitation. The effective radiative forcing associated with aerosol-cloud interactions (ERFaci) reduces from -1.37 to between -0.94 and -1.23 W m-2, depending on the size of giant aerosol. Our results highlight the importance of a better representation of giant aerosol in Earth system models to provide better predictions of cloud, precipitation, and the climate.
- Preprint
(10478 KB) - Metadata XML
-
Supplement
(1558 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-523', Anonymous Referee #1, 29 Mar 2024
This manuscript addresses the treatment of giant aerosols in climate models and performs a number of tests to gauge how this treatment affects cloud/rain fields and ultimately shortwave radiation.
I am sorry to say that I don’t think this is a scientifically well-founded paper. Also, it is so poorly written that it shouldn’t have been let out the door of the lab.
Science:
It’s hard to find justification for this study. The authors know that giant aerosols are hard to measure – they exist in tiny concentrations and don’t pass through inlets easily – and are hard to treat microphysically. The authors modify treatments of these aerosols in a model that doesn’t resolve clouds and parameterizes them with limited success. Aerosol-cloud interactions are certainly not adequately represented in this and all climate models. The microphysics is two-moment but the tails of the distribution are not properly represented. What can one learn from adding giant aerosols in tiny concentrations to these parameterized clouds? In my mind, only that there are so many things wrong about aerosol-cloud interactions in climate models that I wouldn’t start by looking at the tiny tail of the aerosol size distribution as a means of improving agreement with other studies/observations.
A series of tests that short circuit activation microphysics for giant aerosols and sends them to the rain category gives you what you expect. Less cloud, more rain.
In addition, Posselt and Lohmann (2008) already did much of this so I don’t see the motivation.
A few statements are particularly annoying:“Incorporating GCCN results in precipitation enhancement at higher coarse mode aerosol concentration. Our results show that the directly-activated GCCN mechanism produces simulations that are consistent with the LES results Dziekan et al. (2021).”
Given the nature of the tests, the first part states the obvious. The second means nothing since these are two completely different kinds of models.
Another:“These results demonstrate that the directly-activated GCCN mehcanism [sic] is important in simulating the response of surface rain flux to coarse mode aerosol concentration, and, hence, reducing the precipitation susceptibility to aerosol.”
How can one logically draw this conclusion given all the uncertainties in this model’s representation of clouds, precipitation and their precipitation susceptibility?
One more:“This feature essentially reduces the precipitation susceptibility to aerosol, which leads to weaker aerosol ERF, bringing in the model in better agreement with state-of-the-art estimates reported in IPCC AR6 (Forster et al., 2021).”
Unless you can show this occurs for the right reasons, any improvement is basically no better than chance.
Presentation:
The manuscript is full of grammatical problems and colloquialisms. It should have been carefully edited internally before being sent out. It reflects very poorly on the team.Citation: https://doi.org/10.5194/egusphere-2024-523-RC1 -
RC2: 'Comment on egusphere-2024-523', Anonymous Referee #2, 15 Apr 2024
The manuscript describes an attempt at simulating the influence of giant aerosols (GA) on clouds in a global climate model (GCM). Simulating GA is already difficult in smaller-scale models, hence it is very challenging in GCMs with their parameterized clouds and simplistic aerosol-cloud interactions. Adding GA to parameterized clouds leads to large uncertainty about their role. Nevertheless, there is value in such a study as it gives the best (although crude) estimate of the role of GA for climate, and previous studies have found that this role can be significant. However, there are several important issues that need to be addressed before the manuscript is published.
Major comments
1. I do not see significant differences in the treatment of GA between this study and Posselt & Lohmann (2008) (PL hereafter). In this manuscript, as in PL, activated GA are moved directly to rain category, skipping cloud category. Authors claim that the improvement over PL is that their model addresses "kinetic limitations" in the growth of droplets formed on GA. From what I understand, this is done by neglecting GA when computing droplet activation. Whether it is an improvement or not is debatable. More importantly, authors show that it makes no difference if kinetic limitations are addressed. Therefore, the authors need to emphasize more how their study improves on PL and how their results compare with those of PL.
2. Kinetic limitations are addressed in an inconsistent manner. On one hand, GA are excluded from activation to cloud droplets. On the other hand, GA are activated directly to rain drops. This seems contradictory.
3. A more detailed comparison with observations is needed. In particular, size distribution and vertical profile of concentration of GA should be compared. Surface rain flux sensitivity to GA concentration should be compared in more detail with observations from Liu et al. (2022).
4. Presentation quality needs to be improved. Sometimes it is difficult to understand what the authors mean (e.g. l. 98, 111, 117, 146, 151, 153).
5. Including GA causes a significant decrease in cloud droplet concentration over the North-Eastern Europe (Fig. 4 e), although GA concentration over that region is low. What is the reason?
6. Dust and sea-salt GA are treated the same. How realistic is this given their different hygroscopicities?
7. It is surprising that all the rain formed on GA evaporates before reaching the surface. A more detailed discussion of this would be welcome.
Minor comments
- Why the GA threshold sizes are different from those in PL?
- Authors often use "droplet size" where it is not clear if they mean radius or diameter.
- "ug" is used instead of "µg".
Citation: https://doi.org/10.5194/egusphere-2024-523-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 246 | 56 | 25 | 327 | 38 | 17 | 15 |
- HTML: 246
- PDF: 56
- XML: 25
- Total: 327
- Supplement: 38
- BibTeX: 17
- EndNote: 15
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1