the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Nov 2023
Aerosol-Induced Closure of Marine Cloud Cells: Enhanced Effects in the Presence of Precipitation
Matthew W. Christensen
Adam C. Varble
Heng Xiao
Jerome D. Fast
Abstract. The Weather Research Forecasting (WRF) V4.2 model is configured within a Lagrangian framework to quantify the impact of aerosols on evolving cloud fields. Simulations employing realistic meteorological boundary conditions are based on 10 case study days offering diverse meteorology during the Aerosol and Cloud Experiments in the Eastern North Atlantic (ACE-ENA). Cloud and aerosol retrievals in observations from aircraft measurements, ground-based Atmosphere Radiation Measurement (ARM) data at Graciosa Island in the Azores, and A-Train and geostationary satellites are in good agreement with the simulations. Higher aerosol concentration leads to suppressed drizzle and increased cloud water content. These changes lead to larger radiative cooling rates at cloud top, enhanced vertical velocity variance, and increased vertical and horizontal wind speed near the base of the lower-tropospheric inversion. As a result, marine cloud cell area expands, narrowing the gap between shallow clouds and increasing cloud optical thickness, liquid water content, and the top-of-atmosphere outgoing shortwave flux. While similar aerosol effects are observed in lightly to non-raining clouds, they tend to be smaller by comparison. These results show a strong link between cloud cell area expansion and the radiative adjustments caused by liquid water path and cloud fraction changes. These adjustments scale by 74 % and 51 %, respectively, relative to the Twomey effect. Given the limitations of traditional global climate model resolutions, addressing mesoscale cloud-state transitions at kilometer-scale resolutions or higher should be of utmost importance in accurately quantifying aerosol radiative forcing.
- Preprint
(22142 KB) - Metadata XML
-
Supplement
(18048 KB) - BibTeX
- EndNote
Matthew W. Christensen et al.
Status: open (until 13 Dec 2023)
-
RC1: 'Comment on egusphere-2023-2416', Anonymous Referee #1, 22 Nov 2023
reply
Please see the attachment.
-
RC2: 'Comment on egusphere-2023-2416', Michael Diamond, 05 Dec 2023
reply
The authors set up Lagrangian nested WRF simulations at convection-permitting resolution for 10 cases of stratocumulus cloud evolution based on the availability of ACE-ENA flight data and find that as they increase aerosol concentration within the simulations, closed cellular cloud structures tend to expand horizontally (and somewhat vertically as well). The resulting adjustments enhancing liquid water path and cloud fraction together more than double the cooling that would result from the Twomey effect alone on average. Overall, the analysis is well done and the paper is interesting and well-written. I believe some additional nuance would be useful, however, particularly clarifying that the adjustments found in the work are not based on the observations and acknowledging the continuing limitations of the horizontal and vertical resolutions. The discussion of the cloud object method and interpretation could also be clarified. I recommend accepting the manuscript pending minor revisions. -MD
General comments:
A) Model versus observational results: The discussion should better clarify that all aerosol effect conclusion are based on model experiments only. There is no attempt made to deduce aerosol relationships from the observations themselves.
B) Resolution: I agree with the comments about the vertical resolution mentioned by reviewer 1 and think this context should be emphasized more when discussing the positive LWP adjustments. The inability of models to properly represent entrainment and thus the mechanism believed to be behind observed negative LWP adjustments in pollution tracks and effusive volcanic plumes (e.g., Malavelle et al. 2017, Toll et al. 2017) has been repeatedly flagged, as the authors know well.
I also think the discussion of horizontal resolution could use a bit more nuance, as the km-scale resolution is well-suited for resolving the cumulus outflow but is still too course to resolve the updrafts well (Atlas et al., 2022, have a nice treatment of this issue, for example).
Malavelle, F. F., et al.: Strong constraints on aerosol-cloud interactions from volcanic eruptions, Nature, 546, 485-491, 10.1038/nature22974, 2017.
Toll, V., Christensen, M., Gassó, S., and Bellouin, N.: Volcano and Ship Tracks Indicate Excessive Aerosol-Induced Cloud Water Increases in a Climate Model, Geophysical Research Letters, 44, 12492– 12500, 10.1002/2017gl075280, 2017.
Atlas, R. L., Bretherton, C. S., Khairoutdinov, M. F., and Blossey, P. N.: Hallett-Mossop Rime Splintering Dims Cumulus Clouds Over the Southern Ocean: New Insight From Nudged Global Storm-Resolving Simulations, AGU Advances, 3, e2021AV000454, https://doi.org/10.1029/2021AV000454, 2022.
C) Cloud objects: I’m having some trouble interpreting the cloud objects. It seems that for higher aerosol cases, separate updrafts with spreading anvils intersect with each other and are considered one cloud object whereas in the cleaner case they would be treated as separate objects. I can see how this might be helpful for thinking about overall cloud fraction, but it seems like it could be misleading if the number of distinct updrafts isn’t changing between pollution cases.
Specific comments:
1. Lines 72-73: More explanation is needed for the “decreasing seasonal cycle” of CDNC result. I’m assuming you mean that aerosol concentrations are lower in winter than summer, but the higher activation fraction winter leads to a suppressed seasonal difference in CDNC?
2. Sections 2.1 and 2.2: There are no aerosol data listed except for the CPC in ACE-ENA. Figure S2 also includes aircraft CCN data that should be mentioned here. More broadly, I’m surprised that the authors don’t take advantage of the additional aerosol measurements available at ENA. You mention repeatedly that the aerosol concentrations at ENA better resemble the “clean” experiment than the control values, and show this for one case in Fig. S2, but it would be easy to show the issue persists during all cases and better quantify the general bias, differences in aerosol/CCN definitions notwithstanding.
3. Lines 98-99: I don’t understand how excluding this data ensures the “results remain sensitive to variation in aerosol concentration.”
4. Line 108: For the MODIS retrievals shown, which channel is used? I’m assuming the default 2.1 µm? Is there any large sensitivity to this choice?
5. Line 172: The issue isn’t just this day, but rather a general bias throughout both seasons, correct?
6. Lines 205-206: Why was this flight chosen as the main case study?
7. Section 4.1: Why is only the case of 7/18/2017 discussed here? I understand wanting the highlight the results with one flight for illustrative purposes, but from the later figures you have results for all of the flights. It would be helpful to establish here that the case isn’t an outlier and that the results are robust across the simulated days.
8. Lines 231-232: I’m having trouble understanding why larger LWP differences between neighboring pixels would justify merging the objects.
9. Lines 234-235: Why use the minimum distance instead of the mean or median?
10. Line 334/Text S2: The transfer function accounts for transmissivity (reflection and absorption), not just reflection.
11. Lines 335-336/Eq 1/Text S2: Since you’re already accounting for the clear-sky atmospheric transmissivity, this should be the surface albedo.
12. Lines 356-363/Table 3: I’m confused about which experiments are being used to calculate the radiative effects. Is it control-clean, or polluted-pristine? I’d imagine the absolute values should differ quite a bit between those (or other) combinations.
13. Line 370: Aren’t the glaciation effects in Christensen et al. (2014) thought to arise from INP, not just CCN? Are there any INP differences in the experiments?
14. Lines 373-374: Is this boilerplate, or do you mean it? The IPCC is fairly happy to ignore ice-phase aerosol effects as likely small, albeit highly uncertain. Do your results suggest that’s a mistake? (I don’t really see that from the paper, but am open to the argument more generally.)
15. Section 4.3.2: The decision to neglect the cloud fraction adjustments should be given higher real estate here as a caveat, especially as the Morrison microphysics doesn’t allow for full positive aerosol-cloud-precipitation feedback cycle as simulated in some LES (e.g., Yamaguchi et al. 2017). This could have a dramatic influence on cloud fraction (e.g., Goren et al. 2019, Diamond et al. 2022).
Yamaguchi, T., Feingold, G., and Kazil, J.: Stratocumulus to Cumulus Transition by Drizzle, Journal of Advances in Modeling Earth Systems, 9, 2333-2349, 10.1002/2017MS001104, 2017.
Goren, T., Kazil, J., Hoffmann, F., Yamaguchi, T., and Feingold, G.: Anthropogenic Air Pollution Delays Marine Stratocumulus Break‐up to Open‐Cells, Geophysical Research Letters, 46, 14135–14144, 10.1029/2019gl085412, 2019.
Diamond, M. S., Saide, P. E., Zuidema, P., Ackerman, A. S., Doherty, S. J., Fridlind, A. M., Gordon, H., Howes, C., Kazil, J., Yamaguchi, T., Zhang, J., Feingold, G., and Wood, R.: Cloud adjustments from large-scale smoke–circulation interactions strongly modulate the southeastern Atlantic stratocumulus-to-cumulus transition, Atmos. Chem. Phys., 22, 12113-12151, 10.5194/acp-22-12113-2022, 2022.
16. Line 453: The “prior observations” phrasing is misleading, as the adjustments in this work are not observational.
17. Lines 462-463: The phrasing here is a bit awkward, as it reads like autoconversion, not the underestimate of autoconversion, delays raindrop formation.
18. Line 471: Do any of the simulations show aerosols closing open cells into closed cells? I don’t think any of the figures shows this clearly. Should it be more like “aerosols expand the width of closed cells”?
19. Lines 481-483: HX isn’t mentioned in the author contributions.
20. Figure 4: I assume the white stars are the object centroids? This should be mentioned in the caption.
21. Figure 9: Why not just show TKE in panel h?
Citation: https://doi.org/10.5194/egusphere-2023-2416-RC2
Matthew W. Christensen et al.
Matthew W. Christensen et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 199 | 67 | 13 | 279 | 12 | 10 | 8 |
- HTML: 199
- PDF: 67
- XML: 13
- Total: 279
- Supplement: 12
- BibTeX: 10
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1