Dear All,
I think that the reason that we are facing a dilemma is simply because the word "salinity" now stands for a variety of things. That is, we have not been as careful with "salinity" as we now realize we should have been over the past 50 or 100 years.
My understanding of oceanographic practice is that between the Knudsen equation of 1901 and the year 1978 any calculated "salinity" quantity would have tried to be Knudsen salinity, no matter if it was calculated via titration or via a conductivity measurement (as occurred in the 1960s and 1970s). Hence it is quite valid to call such an old data point a "Knudsen_Salinity". Note that we do not reserve a name for a particular measurement technique. Hence, for example, the salinity out of a climate model is called "Practical Salinity" to date even though the value did not result from a conductivity measurement. Also a value of "Absolute Salinity" is still absolute salinity no matter whether it was calculated from either of the two recommended methods.
________________________________
SUGGESTED WAY FORWARD FOR CF-metadata
So any new (>=2010) data going into a data base should be labelled as either "Knudsen Salinity", "Practical Salinity" or "Absolute Salinity", and the name "Salinity" should most definitely not be available as an option for any new (>=2010) data going into a data base, whether from a cruise, a mooring or from model output. I think this is quite clear, yes? I think there can be no argument about this, and to do otherwise would just cause confusion. Can you make this happen in CF-metadata Paul? This is essentially what we are being instructed to do by IOC, IAPSO and SCOR, and it makes perfect sense.
But what should we do about the data in data bases that is presently called "salinity". My suggestion now is that we just let this data sit there as is. We leave it up to the individual researcher to interpret what this data is. For example, in our research, Paul Barker and I get this data and interpret it as Knudsen Salinity if it was collected before 1st January 1978, and Practical Salinity if it was collected on or after 1st January 1978.
________________________________
Can we all agree on the last two paragraphs, and can you, Paul Durack and Nan, make this happen? Until we agree on this as a procedure, there is little point in worrying about the exact wording of the definitions of the variables. Can we all get back to each other with what we think about the suggested way forward that I describe in the above two paragraphs?
Roy, I think what I wrote above is possibly the same as what you are suggesting, but I'm not sure what some of the technical words that you use mean:- "deprecate", "Narrowmatches", etc.
With best wishes,
Trevor
-----Original Message-----
From: Lowry, Roy K. [mailto:rkl at bodc.ac.uk]
Sent: Thursday, 6 October 2011 6:28 PM
To: Rainer Feistel; Durack, Paul (CMAR, Hobart); j.m.gregory at reading.ac.uk; McDougall, Trevor (CMAR, Hobart)
Cc: Stephen.Griffies at noaa.gov; Barker, Paul (CMAR, Hobart); King, Brian A.; rich at eos.ubc.ca; CF-metadata at cgd.ucar.edu
Subject: RE: [CF-metadata] new TEOS-10 standard names
Hello Reiner/Trevor,
I have no problem with introducing the term Knudsen salinity for salinities pre-dating PSS-78. A question to Reiner is does the term just apply to salinity by evaporation/titration or is it equally applicable to salinity data obtained by the STDs and CTDs with home-grown conductivity to salinity algorithms in use before PSS-78 brought some order to the world?
Note, that having a term for pre-78 data doesn't necessarily allow us to immediately deprecate the term 'salinity'. There's a lot of data out there marked up with 'salinity'. Anyone any views on how such deprecation should be managed? My vote would be to set up the new terms and modify the 'salinity' definition in the next Standard Names update, map all the new terms as narrowMatches to old one and then deprecate 'salinity' in a future Standard Names update.
Cheers, Roy.
-----Original Message-----
From: Rainer Feistel [mailto:rainer.feistel at io-warnemuende.de]
Sent: 06 October 2011 06:32
To: Paul.Durack at csiro.au; j.m.gregory at reading.ac.uk; Lowry, Roy K.; Trevor.Mcdougall at csiro.au
Cc: Stephen.Griffies at noaa.gov; Paul.Barker at csiro.au; King, Brian A.; rich at eos.ubc.ca; CF-metadata at cgd.ucar.edu
Subject: Re: [CF-metadata] new TEOS-10 standard names
Since the use of any "salinity" was discouraged in favour of chlorinity in
1940,
all archived pre-78 salinities should be Knudsen salinities, originally
calculated
from chlorinity by the Knudsen formula derived from the evaporation
experiments
of Soerensen in 1900.
Knudsen salinity is consistent with PSS-78 at about S = 35 but deviates
systematically for brackish water, see our 2008 salinity paper in DSR.
The term "Knudsen salinity" is well understood, I think.
Rainer
----- Original Message -----
From: <Trevor.Mcdougall at csiro.au>
To: <rkl at bodc.ac.uk>; <j.m.gregory at reading.ac.uk>; <Paul.Durack at csiro.au>
Cc: <CF-metadata at cgd.ucar.edu>; <rich at eos.ubc.ca>; <bak at noc.soton.ac.uk>;
<Paul.Barker at csiro.au>; <rainer.feistel at io-warnemuende.de>;
<Stephen.Griffies at noaa.gov>
Sent: Wednesday, October 05, 2011 11:17 PM
Subject: RE: [CF-metadata] new TEOS-10 standard names
Hello Roy,
You make a good point about the pre 1978 data. Perhaps we need yet
another name, such as "Pre78salinity" to indicate that it was probably
obtained by chemical titration. By introducing such a salinity will have
the advantage of reducing the present ambiguity which calls this older data
by the same name as Practical Salinity, namely "salinity".
Trevor
-----Original Message-----
From: Lowry, Roy K. [mailto:rkl at bodc.ac.uk]
Sent: Wednesday, 5 October 2011 6:53 PM
To: McDougall, Trevor (CMAR, Hobart); j.m.gregory at reading.ac.uk; Durack,
Paul (CMAR, Hobart)
Cc: CF-metadata at cgd.ucar.edu; rich at eos.ubc.ca; King, Brian A.; Barker, Paul
(CMAR, Hobart); rainer.feistel at io-warnemuende.de; Stephen.Griffies at noaa.gov
Subject: RE: [CF-metadata] new TEOS-10 standard names
Hello Trevor,
I totally agree that we should stop using 'salinity' from now on. I also
agree that virtually all post-1983 (not 1980: it took 5 years to get the
1978 Equation of State published by UNESCO) data labelled 'salinity' are in
fact 'practical_salinity'. However, as an oceanographic data centre we have
salinity data going back to the early 1900s and other centres such as ICES
have data going back further than that. These have been determined by a
variety of methodologies but are mostly chemical titrations or a variety of
algorithmic determinations from conductivity that are significantly
different from the PSS-78 scale. Replacing 'salinity' by
'practical_salinity' re-labels these data, which I believe is wrong.
We certainly need to get 'practical_salinity' names in place and alter the
definition for salinity to indicate that it means 'salt content by any
method' with wording to strongly discourage its use for post-1983 data
unless the data are known to be 'non-practical' (which exist: we have some).
We also need to explain to the community that unless they change the labels
on their data that are practical salinity from 'salinity' to
'practical_salinity' then their data will be regarded as useless for many
physical oceanographic applications.
Cheers, Roy.
-----Original Message-----
From: Trevor.Mcdougall at csiro.au [mailto:Trevor.Mcdougall at csiro.au]
Sent: 05 October 2011 00:12
To: Lowry, Roy K.; j.m.gregory at reading.ac.uk; Paul.Durack at csiro.au
Cc: CF-metadata at cgd.ucar.edu; rich at eos.ubc.ca; King, Brian A.;
Paul.Barker at csiro.au; rainer.feistel at io-warnemuende.de;
Stephen.Griffies at noaa.gov
Subject: RE: [CF-metadata] new TEOS-10 standard names
Dear all,
At the risk of repeating ourselves, because there are now (at least)
three different salinities, it is now ambiguous and confusing to call any
salinity "Salinity". The Announcement of TEOS-10 that is now appearing in
all 22 oceanographic journals specifically recommends that the use of the
word "Salinity" cease immediately, and that either the words "Practical
Salinity" or "Absolute Salinity" be used. The reason of course is to
minimise ambiguity.
So this is where the community (including CF-metadata) will have to end
up:- we have been requested to do so by IOC, SCOR and IAPSO. So we may as
well do it now, in my view.
Note that all ocean models run to date have used Practical Salinity as
their "Salinity" variable,, and all equations of state since 1980 have been
in terms of Practical Salinity. So there is no slight-of-hand in calling
these variables "Practical Salinity"; rather it is just being specific as to
what this type of salinity always has been. That is "Practical Salinity" is
simply the long-hand name of what we have been calling "Salinity" for 30
years.
Trevor
-----Original Message-----
From: Lowry, Roy K. [mailto:rkl at bodc.ac.uk]
Sent: Wednesday, 5 October 2011 1:43 AM
To: Jonathan Gregory; Durack, Paul (CMAR, Hobart)
Cc: McDougall, Trevor (CMAR, Hobart); CF-metadata at cgd.ucar.edu;
rich at eos.ubc.ca; King, Brian A.; Barker, Paul (CMAR, Hobart);
rainer.feistel at io-warnemuende.de; Stephen.Griffies at noaa.gov
Subject: RE: [CF-metadata] new TEOS-10 standard names
Dear All,
My feelings on this were (and still are) comfort with the addition of
'practical_salinity' names, but significant discomfort with the replacement
of 'salinity' by 'practical_salinity' through deprecation of 'salinity'.
Cheers, Roy.
-----Original Message-----
From: cf-metadata-bounces at cgd.ucar.edu
[mailto:cf-metadata-bounces at cgd.ucar.edu] On Behalf Of Jonathan Gregory
Sent: 04 October 2011 14:13
To: Paul.Durack at csiro.au
Cc: Trevor.Mcdougall at csiro.au; CF-metadata at cgd.ucar.edu; rich at eos.ubc.ca;
King, Brian A.; Paul.Barker at csiro.au; rainer.feistel at io-warnemuende.de;
Stephen.Griffies at noaa.gov
Subject: Re: [CF-metadata] new TEOS-10 standard names
Dear Paul
Alison (the manager of standard names) hasn't "ruled" yet on their
inclusion,
but I believe that the discussion concluded with no objections to adding
the practical salinity names. It seems safe to assume they will be put in
the
table in due course.
Best wishes
Jonathan
_______________________________________________
CF-metadata mailing list
CF-metadata at cgd.ucar.edu
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
--
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.
Received on Fri Oct 07 2011 - 19:49:16 BST
