Dear @LorenzoCorgnati

`feature_Type` is defined for discrete sampling geometries. That means 
situations where a combination of two or more continuous dimensions has been 
subsampled, and thus replaced with a smaller number of dimensions that index 
the samples. This is what the start of section 9 of the CF convention says:

> Discrete sampling geometry datasets are characterized by a dimensionality 
> that is lower than that of the space-time region that is sampled; discrete 
> sampling geometries are typically "paths" through space-time.

I'm not clear exactly what you mean by "surface", but perhaps it's a horizontal 
surface. Gridded data on a horizontal surface has two independent dimension x 
and y. If these are independently sampled, meaning that there is a point for 
every possible combination of x and y, it's not a discrete sampling geometry, 
so it doesn't need a `featureType`. You can see in Table 9.1 that all the 
existing `featureTypes` are defined by having a particular combination of 
subsampled dimensions. A distinct new `featureType` would be defined in a 
similar way, by a combinaton of dimensions not so far represented in the table.

If the idea is to use the `featureType` to indicate that the variable *isn't* a 
discrete sampling geometry, as @row-lowry suggests, then something more general 
than `surface` would be appropriate. For example, it could be permitted to 
specify `featureType` as a null string, to indicate that it isn't a DSG, and 
that would be equivalent to omitting the attribute altogether, if others agree 
that there is a general need for this.

Jonathan


-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/cf-convention/cf-conventions/issues/299#issuecomment-693540587

This list forwards relevant notifications from Github.  It is distinct from 
[email protected], although if you do nothing, a subscription to the 
UCAR list will result in a subscription to this list.
To unsubscribe from this list only, send a message to 
[email protected].

Reply via email to