AJ Payne, Geographical Sciences wrote:
> dear all
>
> agreed ice shelves are clearly more closely connected to land ice than
> other elements of the earth system (certainly not sea ice).
>
> i think the cover type mask in glimmer-cism needs to be extended in
> order to make the application of higher order models easier. as a
> minimum:
>
> open ocean
> ice-free land
> ice covered land
> - slow flowing ice
> - ice streams/outlet glaciers
> ice shelves
While it's often handy to have a mask within your code, and to look at
model output to know what the model is doing with different regions, I
think the mask may lock us into the current (esp. Glimmer) way of doing
things. Distinguishing between slow ice and ice streams is very
unnatural since it will always be subjective and some models (e.g. PISM)
do not normally distinguish in this way (solving unified physics
everywhere).
As for grounding line resolution, the combination of bathymetry and ice
thickness provides subgrid resolution where as the mask cannot. What
does the mask provide that cannot be reconstructed with an NCO or
visualization filter?
I'm not objecting to internal use by Glimmer, just to putting this into
CF which may oblige other models to recognize it and obey semantics that
don't make sense in their context.
As for the original topic on cf-metadata_at_ (which I just read from the
archives), I agree that "land ice" should include shelves and I don't
think this definition will cause a problem for users.
Jed
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 261 bytes
Desc: OpenPGP digital signature
URL: <
http://mailman.cgd.ucar.edu/pipermail/cf-metadata/attachments/20090930/56f8e8ac/attachment-0002.bin>
Received on Wed Sep 30 2009 - 03:30:57 BST
