the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Nov 2024
Radiative and climate effects of aerosol scattering in long-wave radiation based on global climate modeling
Abstract. The few studies that considered aerosol scattering in the long-wave (LW) typically relied on artificially increasing it. In order to analyze the radiative and climatic effects of physically accounting for this process, simulations have been performed with the ARPEGE-Climat atmospheric global climate model over the 1985–2014 period, using the ecRad radiation scheme, and updated optical properties of coarse aerosols, particularly dust. The evaluation of the model coarse aerosol optical depth (AOD) against AERONET data over North Africa and Arabian Peninsula shows the ability of ARPEGE-Climat to capture spatio-temporal variations of coarse AOD, despite regional biases. The comparison of simulations with and without aerosol scattering in the LW shows that this process leads to a significant increase in downwelling surface LW radiation in dust-emitting regions (+5 W m-2 on average) between March and September, in line with the maximum coarse AOD. This increase results in a rise in minimum near-surface temperatures of up to +1 °C. It is also associated with an outgoing LW radiation decrease at the top of the atmosphere (TOA). However, during certain months and regions, near-surface temperatures can be significantly reduced due to short-wave surface radiation decreases related to increases in low-level clouds. A precipitation increase over Sahel during September linked to wetter atmospheric layers is also simulated. Neglecting LW aerosol scattering in climate simulations has therefore significant impacts on climate, notably in dust-emitting regions. Globally, the LW aerosol scattering contribution to radiation is of 0.4 W m-2 at both surface and TOA.
- Preprint
(26570 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2024-3659', Anonymous Referee #1, 18 Dec 2024
-
RC2: 'Comment on egusphere-2024-3659', Anonymous Referee #2, 12 Feb 2025
"Radiative and climate effects of aerosol scattering in long-wave radiation based on global climate modeling" Thomas Drugé et al.
This paper is an important advance in providing better radiative transfer schemes in climate models. The redo of ARPEGE-Climate with longwave thermal IR scattering for large dust aerosols represents new work and is timely. The authors did a very nice job in presenting overall, with only minor English language issues. The authors can clean up some of the language issues and clarify the issue of monthly climatologies. The other problem is with the diagnostics that remove clouds ('clear sky') from a self-consistent climate model but assume that surface temperatures and water vapor are unchanged. These diagnostics should be removed or justified.
Overall the paper is well written, an important contribution, and nearly ready to publish.
L35-48: This discussion is important here and should do a better job of helping the reader understand the basic physics of the atmosphere and radiative transfer:
the 8-12 micron window is so well known because H2O and other gases blanket the longer wavelengths and shorter wavelengths; it has nothing to do with the aerosol properties.
it would be nice to discuss what size aerosols need to be to affect this IR window (> 2 microns?), including a discussion of their Q vs effective radius here.
L119: Since you do not calculate the scattering phase function for the aerosols, you are limited to RRTMG's two-stream scattering, which causes heating biases (at least in the uv-vis) over bright albedo regions (Hsu and Prather, 2021, Assessing uncertainties and approximations in solar heating
of the climate system. JAMES, 13, e2020MS002131. doi: 10.1029/2020MS002131). Interestingly enough Hsu & Prather did not do LW scattering but only solar. There is nothing that can be done about the use of 2-stream here alas. The other 4+ stream RT codes are at GFDL and CCC.
L153: I am confused as to whether you used AERONET to evaluate the monthly climatology used in your LWAS calculations or to evaluate the daily TACTIC results used to get the monthly means – that would make more sense, then you can use daylight only model data. Can you please evaluate the issues of diurnal aliasing with your TACTIC model? ARPEGE-Climat is using a climatology – right? Also, is the monthly climatology different for each year?
L162-165: This algorithm is a bit confusing, please make it clear what was done.
L172. Since dust, esp. large dust, is non-spherical, please comment on the errors resulting form Mie assumption as opposed to using T-matrix or PingYang's ice crystal codes to get the scattering phase function.
L173: Glad to see new work on the RI of aerosols.
L218: I thought that the ARPEGE-Climat model being used here for LW scattering is just using an aerosol pre-calculated climatology – what does this r value mean – it should be daily, not monthly means I would expect. If you are evaluating the TACTIC run to generate the climatology, then this discussion needs to be revised.
L238: "several meteorological fields" You have only one or two meteorological fields (with and without LWAS). Maybe you mean diagnostics? or quantities?
L243: These 4 months "cover" or "span" the period of max AOD, they do not "correspond" to it – the latter implies that they are the four months of max AOD.
L245: "clear-sky" conditions are artificial and only in a model, what is the purpose here? You should focus on "all-sky" conditions. Also you need the full climate response in ARPEGE since the surface T is changing. This discussion is odd – who cares if it is more significant in clear-sky? We are talking climate and monthly mean aerosols!
L255: Here is the climate response – excellent.
L260ff: Very interesting climatic shift due to the LWAS, how significant is the change in high cloud (Fig 5) or convective rain – I did not notice any discussion of ensembles or climate variability?
L304: How robust are these differences to issues of climate variability, and would having an interactive aerosol calculation (vs. monthly mean) change these results? e.g. large dust aerosol loading depends on wind shear, lack of cirrus, etc? Also the wind/rain biases in L334.
L313: Does ecRad actually do 3D RT? (pardon my ignorance here)
Table 1. The extensive comparison of All-sky vs Clear-sky is becoming too artificial. For SW, this is fine because water vapor does not affect much. Clouds are clearly correlated with the water vapor and hence just removing clouds without including their effect on water is misleading.
Figure 3ff. A lot of the colored signal here appears to be statistically not significant w.r.t. climate noise. There are some hatched areas, but most of your square are not. For Figure 4, the radiation is clearly significant, but the T is not.
One odd question: does you model include LW scattering by cirrus and stratus? Can you comment on that impact (which would seem to be bigger issue than dust?
Please finish the Dufresne 2002 reference by adding: 'doi: 10.1175/1520-0469(2002)059<1959:LSEOMA>2.0.CO;21, 2002'
Citation: https://doi.org/10.5194/egusphere-2024-3659-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 172 | 41 | 10 | 223 | 11 | 9 |
- HTML: 172
- PDF: 41
- XML: 10
- Total: 223
- BibTeX: 11
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1