[CF-metadata] Surface air temperature, etc.

[Curtis Covey]

In my efforts to make data from the Coupled Model Intercomparison Project
(CMIP) consistent with the CF convention, I've noticed that there 
appears to be
no standard name for "surface air temperature."  Though this quantity has an
unusual definition, involving non standardized and sometimes inconsistent
heights above the surface, it is typical output from most atmosphere and
coupled ocean-atmosphere models.

Should surface air temperature (units: K) be added as a new standard name in
the CF list?  Or is it already there in some form I've missed?  A 
related question
that I will need to face is defining daily max and min values for 
surface air
temperature.  Should these be added to the CF list as separate items, or 
is there
another procedure for dealing with them?

Finally, an unrelated question.  I notice no sea ice concentration 
variable in the
CF list but do notice sea_ice_extent with units of square meters.  Are 
these units
really what was intended?  If so, what exactly does sea_ice_extent mean 
and how
does it relate to sea ice concentration (a percentage or dimensionless 
fraction)?

Curt Covey
Program for Climate Model Diagnosis & Intercomparison