[CF-metadata] New standard names for NEMO ocean model output

From: Jonathan Gregory <j.m.gregory>
Date: Tue, 13 Dec 2016 13:56:41 +0000

Dear Kevin

> 1. bottom_pressure_equivalent_height (m)
> 3. ocean_steric_height (m)
>
> The steric height is estimated as the vertical integral of the density (relative to a reference density where T=0K and S=35psu). The bottom pressure is the mass of the water column at a given location.

Ah, I see.

For
> 3. ocean_steric_height_above_sea_level (m)
I would suggest
> The ocean steric height above sea level measures the change in thickness of a column of water when its temperature and salinity are changed from standard values of 0?C and 0.035 to the actual values

> The bottom pressure equivalent height is estimated indirectly as the difference between the steric height and the sea level.

I don't follow that, which sounds like the definition of ocean steric height
again. However your alternative statement of its being the mass of the column
makes sense to me. Going with the latter definition, I would suggest

sea_water_mass_per_unit_area_expressed_as_thickness

and presumably you have to state a standard density to be used in this
conversion - what is that? NB sea_water_mass_per_unit_area (kg m-2) is already
a standard name.

> 2. Instead of "ocean_turbocline_depth (m)" we suggest:
> ocean_turbocline_thickness (m)
> 'The turbocline thickness is similar to the mixed layer thickness but is estimated in models as the thickness at which the vertical eddy diffusivity coefficient (resulting from the vertical physics alone) falls below a given value defined locally.'

Is there a difference between that and the existing
ocean_mixed_layer_thickness_defined_by_vertical_tracer_diffusivity
?

> ratio_of_sea_water_potential_temperature_anomaly_to_relaxation_timescale (K s-1)
>
> 'This term is estimated as the deviation of the local sea water potential temperature from an ocean model wrt an observation-based climatology (eg World Ocean Database) weighted by a user-specified relaxation coefficient in s-1 (1/(relaxation timescale)). The relaxation coefficient depends on the timescale on which the correction is applied.'

It seems to me that the last sentence is probably not necessary, since the
previous sentence says the same.

Best wishes and thanks

Jonathan
Received on Tue Dec 13 2016 - 06:56:41 GMT