Dear Stephen,
Thank you for getting back to me.
A CF standard name for integral_wrt_depth need not necessarily be interpreted as a full ocean depth quantity. The limits of the integral are specified by placing bounds on the vertical coordinate variablle that is attached to the data variable. The bounds can be used to indicate that an integral has been calculated over a single model layer, for example. We recently discussed on the mailing list how to specify the limits if the integral is calculated over the whole ocean depth and it was agreed that if no limits (i.e. bounds) are specified then the integral is assumed to be full depth. This clarification has now been added to the definitions of all the integral_ wrt_depth standard names.
I think Martin's question regarding the units is an important one and it would be better to be clear in the names that the quantities are vertical integrals if that is indeed the case. The bounds can then be used to describe the limits, as above. I think it would then be okay to describe something as being the tendency of an integrated quantity. My own question related to the order in which the operations are carried out on the variable, I.e. is it the tendency of the vertical integral, or the vertical integral of the tendency?
Best wishes,
Alison
________________________________
From: Stephen Griffies - NOAA Federal <stephen.griffies at noaa.gov>
Sent: 21 June 2018 14:00:46
To: Pamment, Alison (STFC,RAL,RALSP)
Cc: Juckes, Martin (STFC,RAL,RALSP); cf-metadata at cgd.ucar.edu; Jonathan Gregory; Karl Taylor; Durack, Paul J.
Subject: Re: tendency_of_sea_water_conservative_temperature_expressed_as_heat_content units
Alison,
Thanks for staying on top of these matters.
The diagnostics "tendency_of_sea_water_" refer to the tendency as integrated over the thickness of a single model grid cell.
In contrast, "integral_wrt_depth_of_sea_ice_temperature_expressed_as_heat_content " is an integral over the full ocean depth from bottom to top.
I recommend we keep the naming convention unchanged in order to clearly distinguish between the two diagnostics.
Stephen
On Thu, Jun 21, 2018 at 2:42 AM, Alison Pamment - UKRI STFC <alison.pamment at stfc.ac.uk<mailto:alison.pamment at stfc.ac.uk>> wrote:
Dear Martin, Stephen and Jonathan,
We have seven existing tendency_of_sea_water_conservative_temperature_expressed_as_heat_content names (and seven existing tendency_of_sea_water_potential_temperature_expressed_as_heat_content names)all with units of W m-2. I think all of these were introduced for OMIP.
If something is described as a 'heat content' I would expect it to have units of J m-2. Indeed that is the case for the two existing names
integral_wrt_depth_of_sea_ice_temperature_expressed_as_heat_content and integral_wrt_depth_of_sea_water_potential_temperature_expressed_as_heat_content. Calculating tendencies of such quantities would then give us units of W m-2. This suggests to me that the OMIP names should all follow the pattern:
tendency_of_integral_wrt_depth_of_sea_water_X_temperature_expressed_as_heat_content
where X is 'potential' or 'conservative'. The bounds of the vertical coordinate variable should give the limits on the integral for each grid cell.
Does this pattern of writing the names match the method of calculating the quantities (i.e. the tendency of the integral, rather than the integral of the tendency?)
We can of course create aliases to correct the names once we have agreed on what changes are needed.
Best wishes,
Alison
------
Alison Pamment Tel: +44 1235 778065
NCAS/Centre for Environmental Data Archival Email: alison.pamment at stfc.ac.uk<mailto:alison.pamment at stfc.ac.uk>
STFC Rutherford Appleton Laboratory
R25, 2.22
Harwell Oxford, Didcot, OX11 0QX, U.K.
From: Juckes, Martin (STFC,RAL,RALSP)
Sent: 10 June 2018 18:52
To: Stephen Griffies - NOAA Federal <stephen.griffies at noaa.gov<mailto:stephen.griffies at noaa.gov>>
Cc: cf-metadata at cgd.ucar.edu<mailto:cf-metadata at cgd.ucar.edu>; Jonathan Gregory <j.m.gregory at reading.ac.uk<mailto:j.m.gregory at reading.ac.uk>>; Karl Taylor <taylor13 at llnl.gov<mailto:taylor13 at llnl.gov>>; Pamment, Alison (STFC,RAL,RALSP) <alison.pamment at stfc.ac.uk<mailto:alison.pamment at stfc.ac.uk>>; Durack, Paul J. <durack1 at llnl.gov<mailto:durack1 at llnl.gov>>
Subject: Re: tendency_of_sea_water_conservative_temperature_expressed_as_heat_content units
Hi Stephen,
thanks, that is clear. There may be an issue with the CF standard name ... we usually have "integral_wrt_depth" in the name for such quantities. Perhaps Jonathan or Alison can comment on that,
regards,
Martin
________________________________________
From: Stephen Griffies - NOAA Federal <mailto:stephen.griffies at noaa.gov<mailto:stephen.griffies at noaa.gov>>
Sent: 10 June 2018 16:52
To: Juckes, Martin (STFC,RAL,RALSP)
Cc: mailto:cf-metadata at cgd.ucar.edu<mailto:cf-metadata at cgd.ucar.edu>; Jonathan Gregory; Karl Taylor; Pamment, Alison (STFC,RAL,RALSP); Durack, Paul J.
Subject: Re: tendency_of_sea_water_conservative_temperature_expressed_as_heat_content units
Hi,
Thanks for the question.
As discussed in Griffies et al (2016), we request heat and salt budgets to be integrated over the thickness of a grid cell. For the heat budget, this thickness weighting then leads to units of W m-2 rather than W m-3.
There is a good reason to ask for the diagnosed budgets to be integrated over the thickness of a grid cell. Namely, most ocean models have time-dependent grid cell thicknesses. So the only way to ensure budgets can be closed with offline diagnostics is to have each model perform the thickness weighting online.
Make sense?
Best,
Stephen
On Sun, Jun 10, 2018 at 9:58 AM, Martin Juckes - UKRI STFC <mailto:martin.juckes at stfc.ac.uk<mailto:martin.juckes at stfc.ac.uk>> wrote:
Dear Jonathan, Stephen, Karl,
I'm puzzled by the units of the CMIP6 variable ocontemptend and teh associated standard name tendency_of_sea_water_conservative_temperature_expressed_as_heat_content -- in the data request and the standard name table respectively with units "W m-2". This is consistent with the Griffies et al 2016 paper on ocean diagnostics and with the discussion on the CF mailing list. However, it is requested as a function of depth, so I would expect to see units of "W m-3" for the tendency of a heat density.
The units "W m-2" are usually used for a surface heat flux. There are a number of variables related to ocontemptend with the same units.
Am I missing something, or should we change the units or the depth dependency?
regards,
Martin
--
Dr. Stephen M. Griffies
NOAA Geophysical Fluid Dynamics Lab
201 Forrestal Road
Princeton, NJ 08542
USA
--
Dr. Stephen M. Griffies
NOAA Geophysical Fluid Dynamics Lab
201 Forrestal Road
Princeton, NJ 08542
USA
Received on Mon Jun 25 2018 - 03:28:28 BST
