the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Oct 2025
Numerical study of dust plume impact on urban thermal comfort
Abstract. In Europe, heatwave conditions associated with southern synoptic flows can sometimes be combined with dust plume events coming from the Saharian desert. The aerosol plume modifies the radiation observed near the ground and consequently other variables such as air temperature. This study focuses on a heatwave and its corresponding plume of Saharian dusts that have reached the Paris region from 15 to 19 June 2022. To investigate the aerosols impact on the thermal comfort, three numerical simulations based on the Meso-NH model are used: one without dust aerosol (C0), one with CAMS dust data (C1) and one with twice the concentration of CAMS dust data (C2). Simulation C1 is validated against observations from the PANAME-Urban experimental campaign. The impact of aerosols on Aerosol Optical Depth and incoming solar radiation at the surface are well reproduced, and the air temperature and boundary layer heights are improved when taking into consideration the aerosols. The presence of a dust plume during the heat wave contributes to reduce solar radiation and air temperature to up to 75 W/m² and 1 K, respectively. At 16:00 UTC in sunny places of urban and suburban areas, it results in a thermal comfort improvement of up to 1 °C. However, in the shade of suburban areas, the increase of diffuse radiation and relative air humidity as well as the decrease of wind speed induced by the dust aerosols counterbalance the air temperature decrease, thus leading to no improvement of the thermal comfort.
- Preprint
(13415 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 04 Feb 2026)
-
RC1: 'Comment on egusphere-2025-4829', Anonymous Referee #1, 17 Dec 2025
reply
-
AC1: 'Reply on RC1', Jérémy Bernard, 22 Dec 2025
reply
We would like to thank the reviewer for the relevant comments on our manuscript. We will definitely integrate them to submitt an updated version.
Below are some answers to the reviewer first main concern. We indeed did not provide enough information on the model and simulation setup description. More details will be found in the revised manuscript.
1a) "It is not clear, if aerosol arrays are modified by the model or only driven by CAMS data"
The wind model modifies the desert dust aerosols distributions by advection. The CAMS data are only used at the start of the simulation to initialize the position of the plume, and on the outer boundaries (for E, W, S and N boundary conditions).
The other aerosols (rural, urban, marine) are not provided by CAMS and come from a climatology.
1b) "Also, it is not written how model works with aerosols, if includes aerosol-radiation feedback (or first aerosol effect) only, or if aerosols impact clouds and precipitation (second aerosol effect) due to micro-physics scheme in the model and how."
In this article, only the first aerosol effect (aerosol-radiation feedback) is activated, even if more options could be used to activate the second aerosol effect. The reason not to activate the second effect in our simulations is that we focus on a heat wave condition and radiative effect of the desert dust aerosols on thermal comfort. There is no significant cloud cover in the domain of interest.
1c) "Finally, starting describing of experimental setup and C0 simulation is problematic, because aerosol set to zero is non-realistic situation (without cloud and precipitation formation), so first rather describe model (including used aerosol climatology) and the experimental setup later."
The simulation setup is probably unclearly presented. Between C0, C1 and C2, only dust aerosols content is different. Rural, urban and marine aerosols contents are the same for all simulations and they are provided by the climatology. Therefore, the C0 simulation do have a quite common aerosol content for the season and is then completely realistic (C0 represents the most usual case in the Paris region, where the climatology do not have desert dust aerosols at all).
The interest to initialize desert dust aerosols from CAMS is to be able to study these rare events when desert dust from the Sahara desert (more than 2000km away) are advected above Paris region. The C1 and C2 simulations only differ from C0 by these desert dust plume, but all 3 simulations have the same climatology for the common types of aerosols (rural, urban, marine) in the region.
Citation: https://doi.org/10.5194/egusphere-2025-4829-AC1
-
AC1: 'Reply on RC1', Jérémy Bernard, 22 Dec 2025
reply
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 198 | 110 | 34 | 342 | 26 | 24 |
- HTML: 198
- PDF: 110
- XML: 34
- Total: 342
- BibTeX: 26
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The presented study have aim to reveal the impact of plume of a desert dust on temperature and thermal comfort during specific few-day meteorological event by model tools. Although the topic could be interesting, the level of manuscript is rather as report. There is nearly no discussion of results, no comparison with other similar studies, even it they surely exist, some figures are superfluous but some of them unclear. There is also lot of excessive statements, which are not well supported by results. The conclusion is wrongly written. The specific major and other comments are listed below. In the current state, the manuscript is not suitable to be published as a research article, it must be significantly improved or submitted again.
Specific major comments:
1/ Description of model – the inclusion of aerosols in the model, described not very clearly in two distinct paragraphs, does not introduce some important notes. It is not clear, if aerosol arrays are modified by the model or only driven by CAMS data. Also, it is not written how model works with aerosols, if includes aerosol-radiation feedback (or first aerosol effect) only, or if aerosols impact clouds and precipitation (second aerosol effect) due to micro-physics scheme in the model and how. Finally, starting describing of experimental setup and C0 simulation is problematic, because aerosol set to zero is non-realistic situation (without cloud and precipitation formation), so first rather describe model (including used aerosol climatology) and the experimental setup later. All mentioned features should be considered as well as all results commented with regard to those features in discussion.
2/ Discussion section – discussion of results is missing, there is no discussion section, but the results section has only few explanation of achieved results and comparisons with other previous studies in different aspects. Also, general differences between urban and rural temperature daily cycle should be taken into consideration.
3/ Dividing of results into sun and shaded is not introduced, defined and motivated in methodology section and it is unclear how it is selected from model simulations.
Other comments and technical correction:
Line 23: I don‘t think that unit format W/m2 is preferable by ACP, moreover in italics. Also hyphen instead of dash and two near braces don’t look well.
Line 53–61: The research gap and also aim of the study would be better explained.
Line 77–78: The sentence have no sense – no aerosols before 17 June?
L 79–83: It would be appropriate to write, why to choose those models for a comparison. E.g. are they operational system for air-quality predictions through Europe with publicly available results?
L 89: The word part-ly divided.
Fig. 3 caption: The last sentence seems to be describing rather Fig. 2
L 102: Why the line is terminated, when model description continues?
L 122: 0.75 degree in lon/lat is not directly 80 km, please improve the sentence.
L 178–179: The statement is excessive, once there are missing data about midnight 17/18 June.
L 202: I would not say ‘slightly’ – the differences between simulated and observed direct / diffusive radiation are significant.
L 208 and 209: The hyphen is not a minus.
L 221–223: The sentence is excessive, not clear and the final suggestion not well documented
L 242–244: The statement is also not well supported by results.
L 255–257: It is better to link to figures than to sections
L 260: Why simulation C2 is not considered?
Fig. 11 + 12: Why there are missing data in plots, when it comes from model simulations?
L 281–283: You should distinguish between relative and absolute humidity. The equation 1 and also Fig. 12 include the relative, which could be affected by temperature only, but then you describe absolute humidity, also no clear to what refers ‘moister boundary layer’.
L 286–288: I don’t understand the meaning of the sentences.
Fig. 13: It is not clear what difference is shown and the reason why it is shown. If it is UTCI offset, I am not sure about relevance of such evaluation, moreover only in shade.
Fig. 14B: The word each has lost.
L 338–341: The sentence is not logical, because AOD is directly connected with radiation impact on the specific wavelength, not with aerosol size.
Conclusion, second paragraph: It seems as discussion section, but statements are often out of paper topic.