[CF-metadata] water level with/without datum
Dear Jeff, Alison et al.
Like Alison, I prefer sea_surface_height to water_level, because it's really
the same geophysical quantity at the coast and far from the coast, isn't it. It
seems awkward to me to say that the satellite altimeter measures SSH, while a
tide gauge measures water level. The altimeter comes right up to the coast (in
principle), so where is the distinction? Also, water_level is rather like
sea_level, which is the name of a special surface i.e. mean sea level, whereas
SSH is the name of time-varying quantity.
I tend to think that geophysically different datums imply different geophysical
quantities and standard names. The reference surface might also be time-
dependent, in a different way (in general, more slowly) from the SSH; it's not
a constant offset. There is some vagueness about this; some of the reference
surfaces might be different estimates of the "same" thing e.g. different geoids
and different reference ellipsoids. In that case we could use the same stdname,
but we ought to use other attributes to be more precise, when necessary. But
I don't think the lowest tide level, the mean high water, the (mean) sea level
etc. should be regarded as the "same" reference surface. They're all
different. Having a dozen different SSH standard names would be OK, I'd say.
As for sea, lake and river - this is a vexed question we have never resolved!
If only there were a simple word which meant any of the three! At the moment,
we only have "sea" names. A possibility is to define sea to mean sea, lake or
river, in CF standard names; obviously that could be confusing. We could
replace "sea" (almost) everywhere with e.g. "slr", which would not be self-
explanatory, but it would force the user to look it up, and so it would not be
confusing. (We do use a few other abbreviations e.g. toa and lwe.) e.g.
sea_water_temperature -> slr_water_temperature. I think that would be a
reasonable solution, but there are probably better ones.
Best wishes
Jonathan
Received on Fri Feb 12 2010 - 11:49:11 GMT
This archive was generated by hypermail 2.3.0
: Tue Sep 13 2022 - 23:02:41 BST