 |
| Titel |
SNOWMIP2: Comparing simulation of the deep Snowball and the nearly deglaciating Snowball among GCMs |
| VerfasserIn |
D. S. Abbot, A. Voigt, M. Branson, D. Pollard, D. Koll, R. T. Pierrehumbert, G. Le Hir, D. Randall, Y. Donnadieu |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250058403
|
|
|
|
|
|
| Zusammenfassung |
| A variety of global climate models (GCMs) has been used to examine the Snowball Earth
Hypothesis for Neoproterozoic glaciations, often with conflicting results. If such models are
to be used to increase theoretical understanding of Neoproterozoic glaciations, the sources of
discrepancies among models must be understood.
A previous SNOWball Model IntercomParison (SNOWMIP) established that even with
ocean dynamics turned off differences in model ice and snow albedo schemes can cause large
differences in the estimate of the CO2 threshold for Snowball initiation. Furthermore, it was
found that important model differences, likely due to atmospheric dynamics and cloud
scheme behavior, remain even when ice and snow albedos are made uniform among
models.
Here we describe a new SNOWball Model IntercomParison (SNOWMIP2) that is
designed to compare GCM simulation of the deep Snowball, at low CO2 (102 ppm),
and the Snowball near deglaciation, at high CO2 (105 ppm=0.1 bar). SNOWMIP2
includes six different GCMs, all of which have been used in previous Snowball Earth
modelling. Each model is run with uniform land glaciers everywhere, a uniform
surface albedo of 0.6, zero aerosols and ozone, the same CO2 concentration, all
other greenhouse gases set to zero, and with the same orbital parameters and solar
constant.
Because of the uniform boundary conditions, the GCMs produce fairly similar climate
simulations, although there are important differences. For example, equatorial temperatures
differ by as much as 15Â K, at a given CO2, between the warmest and coldest models. This
compares to a warming of 3Â K produced in the CAM GCM when the CO2 is increased from
0.1 bar to 0.2 bar. We are able to attribute these differences mainly to cloud behavior. This
means that until we can be more confident in cloud schemes, it will be difficult to use
GCMs to accurately assess whether a Snowball would deglaciate at a particular CO2
value. The models produce qualitatively similar circulations and hydrological cycles,
although they differ in magnitude. This is important for theories of sea glacier flow in a
Snowball. |
|
|
|
|
|
|