 |
| Titel |
Effects of snow grain non-sphericity on climate simulations: Sensitivity tests with the NorESM model |
| VerfasserIn |
Petri Räisänen, Risto Makkonen, Alf Kirkevåg |
| Konferenz |
EGU General Assembly 2017
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 19 (2017) |
| Datensatznummer |
250154402
|
| Publikation (Nr.) |
 EGU/EGU2017-19491.pdf |
|
|
|
|
|
| Zusammenfassung |
| \begin{document}
Snow grains are non-spherical and generally irregular in shape.
Still, in radiative transfer calculations, they are often treated as
spheres. This also applies to the computation of snow albedo in the
Snow, Ice, and Aerosol Radiation (SNICAR) model and in the Los
Alamos sea ice model, version 4 (CICE4), both of which are employed
in the Community Earth System Model and in the Norwegian Earth
System Model (NorESM). In this work, we evaluate the effect of snow
grain shape on climate simulated by NorESM in a slab ocean
configuration of the model. An experiment with spherical snow grains
(SPH) is compared with another (NONSPH) in which the snow shortwave
single-scattering properties are based on a combination of
non-spherical snow grain shapes optimized using measurements of
angular scattering by blowing snow.
The key difference between these treatments is that the asymmetry
parameter is smaller in the non-spherical case ($\approx 0.78$ in
the visible region) than in the spherical case ($\approx 0.89$).
Therefore, for a given snow grain size, the use of
non-spherical snow grains yields a higher snow broadband albedo,
typically by $\approx$0.03. Consequently, considering the spherical case
as the baseline, the use of non-spherical snow grains results in a
negative radiative forcing (RF), with a global-mean
top-of-the-model value of $\approx -0.22\,\mathrm{W\,m^{-2}}$.
Although this global-mean RF is modest, it has a rather substantial
impact on the climate simulated by NoRESM. In particular, the global
annual-mean 2-m air temperature in NONSPH is 1.17$\,\mathrm{K}$
lower than in SPH, with substantially larger differences at high latitudes.
The climatic response is amplified by strong snow and sea ice feedbacks.
It is further found that the difference between NONSPH and
SPH could be largely "tuned away" by adjusting the snow grain size
in the NONSPH experiment by $\approx 70\%$.
The impact of snow grain shape on the radiative effect (RE) of
absorbing aerosols in snow (black carbon and mineral dust) is also
discussed. For an optically thick snowpack with a given snow grain
effective size, the absorbing aerosol RE is smaller for
non-spherical than for spherical snow grains. The reason for this is
that due to the lower asymmetry parameter of the non-spherical snow
grains, solar radiation does not penetrate as deep in snow as in the
case of spherical snow grains. However, in a climate model
simulation, the RE is sensitive to patterns of aerosol deposition and
simulated snow cover. In fact, the global land-area mean
absorbing aerosol RE is larger in the NONSPH than SPH experiment
(0.193 vs. 0.168$\,\mathrm{W\,m^{-2}}$), owing to
later snowmelt in spring.
\end{document} |
|
|
|
|
|
|