[CF-metadata] Standard names for mean sea level change
Lowry, Roy K.
rkl at bodc.ac.uk
Wed Jun 14 07:10:42 MDT 2017
I sent a response to Karl a couple of weeks back on 'mean sea level', which I'll repeat below:
' As an oceanographic data centre we come across the term 'mean' sea level quite a lot. Common understanding of the generic term is that the averaging interval is any period long enough to remove the tidal signal, typically 1-2 years. This is sufficiently precise for most use cases with the obvious exception of long-term (hundreds of years) studies of observed sea level variation. For these there are precisely defined sea level averages at fixed points given specific names such as Ordnance Datum Newlyn (averaged from 1915 to 1921) or Malin Ordnance Datum (averaged from 1960 to 1969).'
Taken as read when I wrote this is that the averaging is based on time!
Note that we already have a set of Standard Names thus:
My understanding of 'global average sea level change' is spatially averaged change of sea level that contains no tidal signal. This would be eliminated by time averaging in observational data, but may be eliminated from model output by simply not including tides in the algorithm. To me these codes make some sense for model output, but would be meaningless for observational data.
I presume Jonathan's proposed names are to describe change in sea level with the tide removed at a single location. Again, I can understand these when applied to model output, but wouldn't be comfortable with their being applied to observed sea level data without mechanisms to specify both time averaging intervals and the datum.
Therefore two conclusions:
1) I would hold back from adding Jonathan's suggestions until there is an actual use case proposed to reduce the chance of their being used for observational data without the necessary additional metadata.
2) If this suggestion is not accepted then shouldn't the syntax of Jonathan's suggestion be rationalised with the existing global average names?
From: CF-metadata [mailto:cf-metadata-bounces at cgd.ucar.edu] On Behalf Of alison.pamment at stfc.ac.uk
Sent: 14 June 2017 12:20
To: cf-metadata at cgd.ucar.edu
Subject: [CF-metadata] Standard names for mean sea level change
Dear Jonathan, All,
In the NEMO discussion thread we have just agreed three new sea surface height change names. At the end of that discussion Jonathan proposed that we should also add three similar names for changes in mean sea level:
> Although I haven't an immediate use-case, I would say it's very likely
> that the corresponding stdnames for mean sea level change will be wanted sometime.
> are quantities which I have often calculated and plotted, for example,
> but not yet archived in CF-netCDF files! If we add those quantities to
> the standard name table now as well, it might avoid people using the
> SSH names when really they mean MSL. That is, I'd propose we also add
> I put [mean] in brackets because I'm not sure whether we've decided to
> include "mean" in MSL names (that's a different discussion). The above
> quantities are not global average; we already have global average names.
The usual practice has been not to include 'mean' in the names but to include ' "Sea level" means mean sea level' in the definitions. There are only fourteen existing sea level names so it wouldn't require a huge number of aliases if we did decide to change them. Personally, I think it would be useful to create those aliases because it further helps to avoid potential confusion between sea_surface_height and mean_sea_level if someone is just looking through the list of names without delving too far into the definitions. Do you agree?
The next question is what do we actually mean when we say 'mean_sea_level'. I had to go back to 2012 in the mailing list archives to find any discussion of this. There was a discussion about standard names for sea level change, originally proposed by Olivier Lauret. In this discussion we established that 'mean sea level' refers to a time mean although the actual time period is not defined (see http://mailman.cgd.ucar.edu/pipermail/cf-metadata/2012/055733.html). At the same time as creating the aliases I think it would be useful to clarify our existing definitions to say: ' "Sea level" means the time mean of sea level at a given location.' Do you agree? Would it be useful to tie down the definition any more precisely than that, for example, would 'mean sea level' generally be regarded as something representative of a year, a decade, a century, or longer, or do we prefer to leave that completely unspecified?
I assume that the three proposed names are 2D fields which describe the variation in (time) mean sea level at each grid point compared to some previous value. I suggest the following definitions:
thermosteric_change_in_mean_sea_level (m) 'Thermosteric sea level change is the part caused by change in density due to change in temperature i.e. thermal expansion. "Sea level" means the time mean of sea level at a given location. Zero sea level change is an arbitrary level.'
'Halosteric sea level change is the part caused by change in density due to change in salinity. "Sea level" means the time mean of sea level at a given location. Zero sea level change is an arbitrary level.'
'Steric sea level change is caused by changes in sea water density due to changes in temperature (thermosteric) and salinity (halosteric). Zero sea level change is an arbitrary level.'
Here I have adopted the looser wording for steric definitions as currently used in the global average sea level change names. I assume that is more appropriate for these quantities than the stricter definition in the sea surface height names. Is this okay?
Alison Pamment Tel: +44 1235 778065
Centre for Environmental Data Analysis Email: alison.pamment at stfc.ac.uk
STFC Rutherford Appleton Laboratory
Harwell Campus, Didcot, OX11 0QX, U.K.
CF-metadata mailing list
CF-metadata at cgd.ucar.edu
This message (and any attachments) is for the recipient only. NERC is subject to the Freedom of Information Act 2000 and the contents of this email and any reply you make may be disclosed by NERC unless it is exempt from release under the Act. Any material supplied to NERC may be stored in an electronic records management system.
More information about the CF-metadata