Sensitivity of low-level clouds and precipitation to anthropogenic aerosol emission in southern West Africa: a DACCIWA case study

Deroubaix, Adrien; Menut, Laurent; Flamant, Cyrille; Knippertz, Peter; Fink, Andreas H.; Batenburg, Anneke; Brito, Joel; Denjean, Cyrielle; Dione, Cheikh; Dupuy, Régis; Hahn, Valerian; Kalthoff, Norbert; Lohou, Fabienne; Schwarzenboeck, Alfons; Siour, Guillaume; Tuccella, Paolo; Voigt, Christiane

doi:https://doi.org/10.5194/acp-22-3251-2022

Articles | Volume 22, issue 5

https://doi.org/10.5194/acp-22-3251-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-22-3251-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 22, issue 5

Research article

|

11 Mar 2022

Research article |

| 11 Mar 2022

Sensitivity of low-level clouds and precipitation to anthropogenic aerosol emission in southern West Africa: a DACCIWA case study

Adrien Deroubaix, Laurent Menut, Cyrille Flamant, Peter Knippertz, Andreas H. Fink, Anneke Batenburg, Joel Brito, Cyrielle Denjean, Cheikh Dione, Régis Dupuy, Valerian Hahn, Norbert Kalthoff, Fabienne Lohou, Alfons Schwarzenboeck, Guillaume Siour, Paolo Tuccella, and Christiane Voigt

Download

Final revised paper (published on 11 Mar 2022)
Preprint (discussion started on 15 Apr 2021)

Interactive discussion

Status: closed

RC1: 'Comment on acp-2020-1306', Anonymous Referee #1, 23 Apr 2021

Review of Deroubaix et al, Sensitivity of low-level clouds and precipitation to anthropogenic aerosol emission in southern West Africa: a DACCIWA case study

DACCIWA case studies in the monsoon season of July 2016 are simulated with WRF-CHIMERE, focusing on aerosol effects on low clouds and precipitation.

The authors have produced a very well-written paper and there are opportunities for valuable scientific insights. I have some minor comments, below, that should be addressed prior to publication. In particular, if the authors can elucidate the processes in their model that are responsible for their results in more detail, in line with my suggestions in bold text below, I think this will be a very useful and well cited study.

Minor comments:

A few more details on model description would be useful:

What is the vertical resolution of the WRF model at the level of the clouds? How does the Thompson and Eidhammer microphysics scheme do aerosol activation? What are the mechanisms by which aerosols can affect cloud lifetime in this scheme?

Page 3 line 29: is the aerosol size distribution a single variable, or are all 10 bins transferred to WRF? Is there a hygroscopicity for each bin, or does “bulk” mean that is just a single number for each grid cell? Why are deliquesced aerosols treated separately? I didn’t find the reference to Tuccella et al very helpful to figure that out. Does deliquesced aerosols refer to aerosols dissolved in cloud water (referred to as cloud-borne aerosols in the MAM aerosol microphysics schemes)? Also, via the coupler, what fields come back from WRF to CHIMERE, to handle aerosol scavenging for example?

Would be better to introduce the models before describing the coupling.

Is there a sub-grid cloud fraction scheme? If so how does it work?

Page 7: I would say “the nitrate and ammonium concentrations are a factor 100 higher” rather than “are multiplied by 100” as “multiplied” implies you fixed these concentrations deliberately, while in fact, if I understand correctly, it is a model result.

Page 11 line 6 “denote” is the wrong word here.

Line 22: “the processes involving supersaturation to create liquid water are not represented in the model” - -might make sense to rephrase – in the model, the RH is always below 100% because clouds form at 100% RH? (presumably the model produces clouds somehow, even if this is via saturation adjustment).

Page 15 line 3: this is just a phrasing issue, but comparing cloud base height to liquid water mixing ratio doesn’t make sense: maybe you compare the cloud base height and LMWR in the model to the observations, or between two time periods?

Page 15 line 8: what processes in the model lead to increased LWP and cloud cover? Are they indirect effects or semi-direct effects? Are there missing processes in the model that could lead to the opposite effects? Like evaporation/entrainment or sedimentation/entrainment feedbacks for example (see e.g. Hill et al (2009), https://journals.ametsoc.org/view/journals/atsc/66/5/2008jas2909.1.xml)?

Figure 8a: would be really nice to put MODIS or AMSR or SEVIRI liquid water path data on this plot, for times when you have the retrievals.

Page 20: did Menut et al 2019 give reasons for the low bias in precipitation? Do you have insights from this study?

Page 22 line 16 and page 23 line 6: I think it is necessary to add a caveat “aerosols emitted from anthropogenic activities have a regional scale influence on LLC and precipitation IN OUR SIMULATIONS, both….” Modeling these aerosol-cloud interactions is not so easy, and there is no guarantee the model is right!

Figure A3, A4 can you add horizontal snapshots of the cloud cover in the simulations at the same times over this area, or a subset of it? Ideally for both AE1 and AE10?

Citation: https://doi.org/10.5194/acp-2020-1306-RC1
RC2: 'Comment on acp-2020-1306', Anonymous Referee #3, 15 Aug 2021

I have reviewd "Sensitivity of low-level clouds and precipitation to anthropogenic aerosol emission in southern West Africa: a DACCIWA case study" by Deroubaix et al. The title succinctly summarizes the study. The study suffers from the problem inherent in case studies, namely generalizability. But it is solid work, and I recommend publication after minor revisions to address my concerns.

My first concern is that the conclusions (precipitation suppression by anthropogenic aerosols delays the breakup of clouds) are mainly a reflection of the cloud physics included in the model. But models are by necessity incomplete. While this model includes a precipitation suppression mechanism via its precipitation microphysics, there are other aerosol effects that could lead to an enhanced loss of cloud cover through evaporation (e.g., Ackerman et al., 2004). These effects are unlikely to be correctly represented in a 5 km resolution model without convection parameterization because the relevant scales are much smaller for shallow clouds. Over all, the enhanced evaporation effect is as strong (Toll et al., 2019) or stronger (Gryspeerdt et al., 2019) than the precipitation suppresion effect, but there is likely to be a great amount of diversity depending on cloud regime, aerosol loading, etc.

To gauge how much to trust a model that only parameterizes the precipitation mechanism, it would be extremely helpful to know whether the breakup of the clouds discussed in this case study is mainly evaporation-driven or precipitation-driven to begin with. This is of course easier said than done, because we don't have observations of evaporation flux. But a good starting point would be to ask the model: what fraction of the LWP tendency can be explained by evaporation and what fraction by precipitation? If precipitation plays a sizable role in the cloud dissipation in the model, then the next question is whether the precipitation timeseries shown in Fig. 10 agrees with observations. Therefore, I was disappointed that Fig. 10 does not include any observations at all. It would also be useful to include more description of these clouds; I assume they are fairly deep (but still warm) cumulus clouds for which precipitation dissipation is a reasonable assumption, but the onus is on the authors to make this argument.

My second concern is representativeness. Recognizing that aerosol effects on LWP and cloud cover tend to be subtle and can have either sign, it is hard to draw a general conclusion from this study, even setting the model correctness concern aside for the moment, and I am struggling to identify anything new that I have learned from reading the manuscript. This is of course a general problem of case studies with no easy solution. Ideally, the manuscript would make connections to other work, e.g., longer time period regional modeling, and discuss how this analysis corroborates or modifies conclusions of those longer-term studies. Another approach would be to perform an ensemble of model runs for this case study to explore how robust the conclusions are to meteorological variability or model physics (depending on how the ensemble is designed). Model runs (especially ensembles) are not free, so I do not expect the authors to come up with additional analysis. However, I think it is important for the authors to at least discuss representativeness in the final paper.

Citation: https://doi.org/10.5194/acp-2020-1306-RC2
RC3: 'Comment on acp-2020-1306', Anonymous Referee #4, 27 Aug 2021

Review of “Sensitivity of low-level clouds and precipitation to anthropogenic

aerosol emission in southern West Africa: a DACCIWA case study, bu Deroubaix et al.

This manuscript presents a nice study on aerosol impact on cloud cover and precipitation in the southern West Africa. To evaluate the effects, the authors use the combined Chimere-WRF model, and first compare the model results against observations. They find that increased aerosol loading has moderate effect on precipitation and cloud cover. The main findings are changes in cloud breakup time and precipitation timing, and with increased aerosol loading leading to slightly reduced precipitation. The paper is well written, clear to understand and I recommend publishing after addressing some minor issues.

Minor comments:

1)    Page 4, line 8: By adding spectral nudging, is it possible that some of the aerosol effect on dynamics is lost? This comes up again on page 13, lines 3-5. I would not expect the aerosols to have a large effect on the rh and wind, but could spectral nudging also reduce any impact (specifically on the wind)? Perhaps you could add a small discussion regarding how spectral nudging impact aerosol indirect effect evaluation.

2)    Page 20, line 9: Could the relative difference between AE1 and AE10 be larger over the ocean because the aerosol loading there is lower (cleaner)? Large changes in aerosols over a clean area is likely to impose a larger effect compared to increase aerosol loading in an already polluted area.

Technical comments:

Page 2, line 14. I would add Thompson and Eidhammer (2014) here as well.

Page 2, line 32-33. This sentence is hard to read. Please rephrase.

Page 3, line 9. Change Inflow to inflow.

Page 4, line 15: Perhaps add Iacono 2008

Page 6, line 8. Why is sulfate not being evaluated?

Page 10, line 7. I would suggest to shortly describe how RH affect aerosol optical properties.

Page 19, line 23. Please refer to Figure 1 here.

Page 22, line1. After 12:00 UTC , you could add “the next day”

Citation: https://doi.org/10.5194/acp-2020-1306-RC3
AC1: 'Comment on acp-2020-1306', Adrien Deroubaix, 08 Nov 2021

Dear Editor,

I am pleased to submit our responses to the three reviewers (in the attached document) whom we warmly thank for all their insightful comments which clearly improved the manuscript.

Best regards,

Adrien Deroubaix on behalf of all the authors

Citation: https://doi.org/10.5194/acp-2020-1306-AC1

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

AR by Adrien Deroubaix on behalf of the Authors (15 Nov 2021) Author's response

EF by Polina Shvedko (15 Nov 2021) Manuscript

EF by Polina Shvedko (15 Nov 2021) Author's tracked changes

ED: Referee Nomination & Report Request started (24 Dec 2021) by Nikos Hatzianastassiou

RR by Anonymous Referee #1 (30 Dec 2021)

ED: Publish subject to technical corrections (01 Feb 2022) by Nikos Hatzianastassiou

AR by Adrien Deroubaix on behalf of the Authors (08 Feb 2022) Manuscript

Short summary

During the summer monsoon in West Africa, pollutants emitted in urbanized areas modify cloud cover and precipitation patterns. We analyze these patterns with the WRF-CHIMERE model, integrating the effects of aerosols on meteorology, based on the numerous observations provided by the Dynamics-Aerosol-Climate-Interactions campaign. This study adds evidence to recent findings that increased pollution levels in West Africa delay the breakup time of low-level clouds and reduce precipitation.