 |
| Titel |
Constraining cloud droplet number concentration in GCMs suppresses the aerosol indirect effect |
| VerfasserIn |
J. E. Kristjánsson, C. Hoose, A. Kirkevåg, T. Iversen, Ø. Seland, A. Gettelman |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250026975
|
|
|
|
|
|
| Zusammenfassung |
| Over the last decade, aerosol-climate models have evolved to include more and more
sophisticated aerosol and cloud microphysics schemes. With a prognostic treatment of
aerosol number concentration and a continuity equation for cloud droplets, very low values
for the droplet concentration can occur, which are not considered realistic. To avoid this,
several state-of-the-art GCMs constrain the droplet concentration such that it can not fall
below a certain minimum value inside a cloud or impose lower bounds for aerosol
concentration. While it can be argued that natural background aerosol particles
(e.g. non-desert dust or bioaerosols) are missing in most GCMs and therefore the
simulated concentrations are too low, observed cloud droplet concentrations can actually
fall below 30 cm-3 in the remote oceans or even below 15 cm-3 in the Arctic.
This suggests that lower bounds are problematic, especially for clean preindustrial
conditions.
It has been shown previously (Lohmann et al, 2000: JGR; Wang & Penner, 2008: ACPD)
that this constraint has the side effect of reducing the simulated aerosol indirect effect. Here
we investigate this effect systematically with the CAM-Oslo GCM (Storelvmo et al, 2006:
JGR; Seland et al, 2008: Tellus). Setting the lower bound for the cloud droplet
concentration to 0, 1, 10, 20 and 40 cm-3 results in changes of the shortwave cloud forcing
between present-day and preindustrial conditions which range from -1.9 Wm-2
(without a lower bound) to -0.6 Wm-2 (lower bound of 40 cm-3). A similar effect is
found if the cloud droplet concentration is treated diagnostically, for which case
the results range from -2.4 Wm-2 (lower bound of 1 cm-3) to -0.8 Wm-2 (lower
bound of 40 cm-3). We will discuss these simulations and the implications of our
findings. |
|
|
|
|
|
|