[Go-essp-tech] Fwd: [Fwd: Variable names for CMIP5] ISSUE FOR FILENAMES

[Bryan Lawrence]

----------  Forwarded Message  ----------

Subject: [Fwd: Variable names for CMIP5]
Date: Tuesday 17 February 2009
From: Alejandro Bodas-Salcedo <alejandro.bodas at metoffice.gov.uk>
To: bryan.lawrence at stfc.ac.uk

-------- Forwarded Message --------
From: Alejandro Bodas-Salcedo <alejandro.bodas at metoffice.gov.uk>
To: Karl Taylor <taylor13 at llnl.gov>
Cc: Gregory, Jonathan <jonathan.gregory at metoffice.gov.uk>, Webb, Mark
<mark.webb at metoffice.gov.uk>, Sandrine Bony
<Sandrine.Bony at lmd.jussieu.fr>, Bentley, Philip
<philip.bentley at metoffice.gov.uk>, alison.pamment at stfc.ac.uk
Subject: Variable names for CMIP5
Date: Mon, 16 Feb 2009 12:41:50 +0000
Dear Karl,

Last week, I proposed new variable names to be included in the table of
CF standard names for the outputs from COSP (there is currently a
discussion going on on the CF-metadata mailing list). Jonathan very
sensibly pointed out that we should name variables according to the
geophysical quantities that they represent, and include the details on
how they are computed as metadata. Then, the question is how to
disambiguate these variables in CFMIP5 (e.g. cloud_area_fraction as
computed from different instrument simulators).

Mark W. and I have thought of two options to disambiguate these
variables:
a) Use different MIP tables for different sources (e.g instrument
simulators). With the CMIP3 naming convention the table name is added as
a suffix, so this should provide a unique filename (clt_A1,clt_A2,etc).
b) Use a suffix with the identification of the source
(clt_A1_calipso,clt_A1_cloudsat,etc).

Option (a) seems to be more practical, as it does not introduce special
cases in the data processing. However, this option might introduce
confusion and I think it may lead to users using the wrong data. Option
(b) is probably better as the source is coded in the filename, reducing
the risk of misinterpration.

The latest draft of the AR5 Data Reference Syntax does not seem to
include the table name as part of the file name (is this the case?),
which will rule out option (a). It also mentions that the 'cell_method'
will be included as part of the filename. This could be used to include
the simulator used to produced the data, without the need of a suffix,
as it would be appended to the standard variable name.

I wonder whether you could comment on the feasibility of these options
in CFMIP5, or suggest the best approach to dealing with this.

Regards,

Alejandro


-- 
------------------------------------------------------------------
Dr Alejandro Bodas-Salcedo     Earth Observation Research Scientist
Met Office Hadley Centre
FitzRoy Rd   Exeter  EX1 3PB   United Kingdom
Tel:  +44 (0)1392 886113   Fax:  +44 (0)1392 885681
E-mail: alejandro.bodas at metoffice.gov.uk   http://www.metoffice.gov.uk

Met Office climate change predictions can now be viewed on Google Earth 
http://www.metoffice.gov.uk/research/hadleycentre/google/
------------------------------------------------------------------

-------------------------------------------------------

-- 
Bryan Lawrence
Director of Environmental Archival and Associated Research
(NCAS/British Atmospheric Data Centre and NCEO/NERC NEODC)
STFC, Rutherford Appleton Laboratory
Phone +44 1235 445012; Fax ... 5848; 
Web: home.badc.rl.ac.uk/lawrence