[CF-metadata] Adding new standard names for satellite altimetry

From: Jonathan Gregory <j.m.gregory>
Date: Wed, 9 Apr 2008 20:17:39 +0100

Dear Olivier

Thanks for your altimeter-related names. Some comments:

> sea_surface_height_correction_due_to_air_pressure_and_wind_at_high_frequency (m)
> sea_surface_height_correction_due_to_air_pressure_at_low_frequency (m)

I think "correction" is fine, myself. To me it would mean something that was
added to the data to obtain the (final) sea surface height. There is a possible
ambiguity that a user might not know whether it had been added already, or
should be added, and the definition would have to clarify that. It would
be good to clarify "high" and "low" frequency in the definition too, if
they can't be specified precisely. If they can be specified precisely you
could consider using a coord variable to do so.

> backscatter_coefficient
Perhaps sea_surface_backscatter_coefficient would be a helpful clarification.
I note that we have got backwards_scattering_coefficient in an existing
standard name. Would that be a possibility? (I think, however, that it should
have been "backward", not "backwards", as "backward" seems much more common in
Google - but that's a trivial point.)

> sea_surface_height_bias_due_to_sea_surface_roughness (m)
OK.

> ionosphere_path_delay (m)
> troposphere_wet_path_delay (m)
> troposphere_dry_path_delay (m)
It strikes me as a bit surprising to have a delay in units of length. Perhaps
these could be called altimeter_range_correction_due_to_X like the first ones?

> range (m), or scanned_pulse_distance (m)
I think altimeter_range would be most informative, since that is what the
measurement is usually called, as you say. There would be no problem in
defining a lidar_range separately. Do you have a need for a generic name
at the moment?

> altitude_above_reference_ellipsoid (m)
That should be height_above, I think, because the standard_name of altitude
means height above the geoid, and height in general means vertical distance.

> sea_surface_height_amplitude_due_to_geocentric_ocean_tide (m)
> sea_surface_height_amplitude_due_to_earth_tide (m)
> sea_surface_height_amplitude_due_to_pole_tide (m)
> sea_surface_height_amplitude_due_to_equilibrium_ocean_tide (m)
> sea_surface_height_amplitude_due_to_non_equilibrium_ocean_tide (m)
> geoid_height_above_reference_ellipsoid (m)
OK (given definitions of the tides).

> mean_sea_surface_height_above_reference_ellipsoid (m)
> mean_sea_surface_height_above_geoid (m)
What does "mean" mean here? Is it a time-mean? If so, that should be indicated
in the cell_methods, not the standard name. We already have names like these
without "mean_" prefixed.

Best wishes

Jonathan
Received on Wed Apr 09 2008 - 13:17:39 BST