<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<font face="Times New Roman">Hi Jamie,<br>
<br>
Nice suggestion. I forgot that this info. is supposed to be in
the catalogs. Would seem to be definitely easier than downloading
files and also easier to automate updates. Don't know if it's
easier than querying the CIM, though.<br>
<br>
Karl<br>
</font><br>
On 2/17/12 10:06 AM, Kettleborough, Jamie wrote:
<blockquote
cite="mid:E51EDFEBF10BE44BB4BDAF5FC2F024B90FB7FB81@EXXMAIL02.desktop.frd.metoffice.com"
type="cite">
<pre wrap="">Hello,
I guess there may also be the question of keeping any summary up to date
as new historicalMisc experiments appear. If you want to do this in
'close to real time' then I think Karl's suggestion of inferring from
the NetCDF files is promising - you *could* imagine something in the
'publish' chain extracting the relevant NetCDF attributes. In fact
aren't they there in the thredds catalogs? For instance:
<a class="moz-txt-link-freetext" href="http://cmip-dn.badc.rl.ac.uk/thredds/esgcet/7/cmip5.output1.MOHC.HadCM3">http://cmip-dn.badc.rl.ac.uk/thredds/esgcet/7/cmip5.output1.MOHC.HadCM3</a>.
historical.day.ocean.day.r2i1p1.v20110728.xml
Has a lines like:
<property name="ensemble" value="r2i1p1"/>
:
<property name="forcing" value="GHG, Oz, SA, Sl, Vl, (GHG = CO2, N2O,
CH4, CFCs)"/>
(I know this isn't a historicalMisc, but I'm running out of time - I
think the principles the same - *provided* all thredds catalogs have
extracted the same meta-data).
So a variant on Karl's suggestion is to gather the information from the
thredds catalogs for the historicalMisc experiments. (we can't quite do
this straight off with our thredds reader as we don't read these
properties - we've not needed them, until maybe now). I don't know
enough about publishing from a data node to know how this would fit in
if you wanted to do it for all new historicalMisc experiments as they
were published.
Clearly both NetCDF (and by implication the thredds properties) and CIM
metadata are prone to human error - figuring out what to put in the
questionnaire or what values to pass to CMOR isn't straight forward.
I'm not sure if one is any more prone to error than the other. I
suppose if you were really keen you could try both ways and see how much
things match up, are there any obvious inconsistencies?
Sorry I think I started on one train of thought, then got sidetracked
onto another - I hope the general ideas are useful though, even if I've
not expressed myself as clearly as I should have.
Jamie
</pre>
<blockquote type="cite">
<pre wrap="">-----Original Message-----
From: <a class="moz-txt-link-abbreviated" href="mailto:go-essp-tech-bounces@ucar.edu">go-essp-tech-bounces@ucar.edu</a>
[<a class="moz-txt-link-freetext" href="mailto:go-essp-tech-bounces@ucar.edu">mailto:go-essp-tech-bounces@ucar.edu</a>] On Behalf Of Bryan Lawrence
Sent: 17 February 2012 16:51
To: <a class="moz-txt-link-abbreviated" href="mailto:go-essp-tech@ucar.edu">go-essp-tech@ucar.edu</a>
Cc: Eric Guilyardi
Subject: Re: [Go-essp-tech] How does "experiment=volcanic
aerosol" come tobe in CMIP5?
Hi Karl
One might get more information from a sql query on the CIM
database, or get it easier.
I'd discuss it with Eric and Gerry before you did anything.
Cheers
Bryan
</pre>
<blockquote type="cite">
<pre wrap="">Bryan & Balaji,
I'm all for what's straightforward. I'm also for putting
</pre>
</blockquote>
<pre wrap="">together a
</pre>
<blockquote type="cite">
<pre wrap="">summary of this information sooner, rather than later. I think the
IPCC authors already put together a table for the normal
</pre>
</blockquote>
<pre wrap="">"historical" runs.
</pre>
<blockquote type="cite">
<pre wrap="">Couldn't we just download one file from each of the historicalMisc
runs, and write a script to read what's in the global
</pre>
</blockquote>
<pre wrap="">"forcing" attribute?
</pre>
<blockquote type="cite">
<pre wrap="">We'd create a 3-column table with model name, p-value, and
</pre>
</blockquote>
<pre wrap="">the list of
</pre>
<blockquote type="cite">
<pre wrap="">forcings. This would take someone a few hours, of course,
</pre>
</blockquote>
<pre wrap="">and would
</pre>
<blockquote type="cite">
<pre wrap="">assume folks had correctly recorded the forcings in the
</pre>
</blockquote>
<pre wrap="">attribute, but
</pre>
<blockquote type="cite">
<pre wrap="">we could do this now, and not wait for everyone to complete the
metafor questionnaire.
On the other hand, perhaps the documentation will become available
soon enough not to warrant the extra work. what do you think?
thanks,
Karl
On 2/17/12 8:21 AM, V Balaji wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Agreed!
I think the effort Karl is proposing to undertake (to
</pre>
</blockquote>
</blockquote>
<pre wrap="">tabulate the
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">forcings associated with each p-value for each model) is a
monumental one without CIM metadata, and reasonably
</pre>
</blockquote>
</blockquote>
<pre wrap="">straightforward
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">using the CIM.
On Fri, Feb 17, 2012 at 3:32 AM, Bryan Lawrence
<a class="moz-txt-link-rfc2396E" href="mailto:bryan.lawrence@ncas.ac.uk"><bryan.lawrence@ncas.ac.uk></a> wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Hi Gary
This is precisely what the metadata side of things is
</pre>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">supposed to pick up. Future versions of the user interface
will have links between data and metadata, so you can click
through to the distinguishing factors, which hopefully will
have been entered in the simulation descriptions.
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">
Cheers
Bryan
</pre>
<blockquote type="cite">
<pre wrap="">Hi Gary,
To be sure, the experiment_id for the CCCMA dataset shown below
was (if
correct) "historicalMisc". When they published it,
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">they chose to
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">name the experiment "volcanic aerosol". This is
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">informational but
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">is not used by the ESG search engines or in any other automatic
processing that I know of. (I thought whoever was publishing
would just put in experiment="historicalMisc", so it would be
consistent with the experiment_id global attribute.)
I don't think we've provided much guidance about this
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">(outside the
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">DRS document), but for each different combination of
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">forcings in
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">the historicalMisc runs, you should assign a different
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">"physics_version"
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">number (i.e., the "p" in the ensemble identifier, which in the
example below is "p3"). You should also record in the
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">global "forcing"
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">attribute in the netCDF file each of the forcings
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">included in each
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">of your historical(Misc) runs. If you do an ensemble
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">of runs with
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">the same forcing, they should all have the same "p"
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">value and be
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">distinguished by the "r" values. The standard
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">abbreviations for
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">each forcing are provided in the last appendix of the
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">DRS document
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">and also can be found at
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap=""><a class="moz-txt-link-freetext" href="http://cmip-pcmdi.llnl.gov/cmip5/docs/cmip5_data_reference_Appendi">http://cmip-pcmdi.llnl.gov/cmip5/docs/cmip5_data_reference_Appendi</a>
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">x1-2.pdf [Note that the output_requirements, available at
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap=""><a class="moz-txt-link-freetext" href="http://cmip-pcmdi.llnl.gov/cmip5/docs/CMIP5_output_metadata_requir">http://cmip-pcmdi.llnl.gov/cmip5/docs/CMIP5_output_metadata_requir</a>
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">ements.pdf , describes the "forcing" attribute.]
I would like to put together a table that provides a
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">key showing
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">which forcings are included for each model and for each
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">value of
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">"p"; I just haven't gotten to this yet.
Please let me know if you can think of any way to make this
clearer to everyone.
Best regards,
Karl
On 2/16/12 7:31 AM, Gary Strand wrote:
</pre>
<blockquote type="cite">
<pre wrap="">At what point (publishing?) does one of an "historicalMisc"
experiment (as defined and required by the CMIP5
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">tables) get displayed as:
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">
I would like our (NCAR) "historicalMisc" datasets have
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">a bit more
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">information available than just the standard (as
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">described in the
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">DRS
document) "historical simulation but with other individual
forcing agents or combinations of forcings."
We have a whole slew of variations that are not readily
distinguishable (and less likely usable) by the above
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">generic description.
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">
Thanks!
Gary Strand
<a class="moz-txt-link-abbreviated" href="mailto:strandwg@ucar.edu">strandwg@ucar.edu</a>
_______________________________________________
GO-ESSP-TECH mailing list
<a class="moz-txt-link-abbreviated" href="mailto:GO-ESSP-TECH@ucar.edu">GO-ESSP-TECH@ucar.edu</a>
<a class="moz-txt-link-freetext" href="http://mailman.ucar.edu/mailman/listinfo/go-essp-tech">http://mailman.ucar.edu/mailman/listinfo/go-essp-tech</a>
</pre>
</blockquote>
</blockquote>
<pre wrap="">--
Bryan Lawrence
University of Reading: Professor of Weather and Climate
</pre>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">Computing.
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">National Centre for Atmospheric Science: Director of
</pre>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">Models and Data.
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">STFC: Director of the Centre for Environmental Data Archival.
Ph: +44 118 3786507 or 1235 445012;
</pre>
</blockquote>
</blockquote>
</blockquote>
<pre wrap="">Web:home.badc.rl.ac.uk/lawrence
</pre>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">_______________________________________________
GO-ESSP-TECH mailing list
<a class="moz-txt-link-abbreviated" href="mailto:GO-ESSP-TECH@ucar.edu">GO-ESSP-TECH@ucar.edu</a>
<a class="moz-txt-link-freetext" href="http://mailman.ucar.edu/mailman/listinfo/go-essp-tech">http://mailman.ucar.edu/mailman/listinfo/go-essp-tech</a>
</pre>
</blockquote>
<pre wrap="">
</pre>
</blockquote>
<pre wrap="">
</pre>
</blockquote>
<pre wrap="">
--
Bryan Lawrence
University of Reading: Professor of Weather and Climate Computing.
National Centre for Atmospheric Science: Director of Models and Data.
STFC: Director of the Centre for Environmental Data Archival.
Ph: +44 118 3786507 or 1235 445012;
Web:home.badc.rl.ac.uk/lawrence
_______________________________________________
GO-ESSP-TECH mailing list
<a class="moz-txt-link-abbreviated" href="mailto:GO-ESSP-TECH@ucar.edu">GO-ESSP-TECH@ucar.edu</a>
<a class="moz-txt-link-freetext" href="http://mailman.ucar.edu/mailman/listinfo/go-essp-tech">http://mailman.ucar.edu/mailman/listinfo/go-essp-tech</a>
</pre>
</blockquote>
<pre wrap="">_______________________________________________
GO-ESSP-TECH mailing list
<a class="moz-txt-link-abbreviated" href="mailto:GO-ESSP-TECH@ucar.edu">GO-ESSP-TECH@ucar.edu</a>
<a class="moz-txt-link-freetext" href="http://mailman.ucar.edu/mailman/listinfo/go-essp-tech">http://mailman.ucar.edu/mailman/listinfo/go-essp-tech</a>
</pre>
</blockquote>
</body>
</html>