Dear Jonathan,
That sounds reasonable, but does it deal with the possibility that there is more than one 0 degC isotherm? If the ground is unfrozen and there is one or more 0 degC isotherms below ground, we want the depth to the first one. Would "cell_methods = depth: minimum" deal with this? The usage of asking for a depth minimised over depth looks a little odd to me, but perhaps it is OK. regards, Martin ________________________________ From: CF-metadata <[email protected]> on behalf of Jonathan Gregory <[email protected]> Sent: 18 May 2018 13:48 To: [email protected] Subject: Re: [CF-metadata] Standard names for LS3MIP: 8 temporal changes + 1 feature depth Dear Martin Ah, I see. Sorry, I hadn't grasped that point. I certainly don't want to try to persuade LS3MIP to provide something to suit CF. CF should try to describe what people want to use. I think your suggestion below is a good one. We could use the standard_name of depth with a coordinate variable soil_temperature=0 degC and a cell_methods specification "where unfrozen_ground", for example. We already use "ground" in the phrase at_ground_level in several standard names; it means the top of the soil or rock surface, underneath the snow and ice, if any. "ground" is in some other names too, and the bare_ground area_type. "unfrozen" is in the guidelines and in one extant name. Best wishes Jonathan ----- Forwarded message from Martin Juckes - UKRI STFC <[email protected]> ----- > Date: Fri, 18 May 2018 08:29:34 +0000 > From: Martin Juckes - UKRI STFC <[email protected]> > To: "[email protected]" <[email protected]>, > "[email protected]" <[email protected]> > Subject: Re: [CF-metadata] Standard names for LS3MIP: 8 temporal changes + 1 > feature depth > > Dear Jonathan, > > > I agree that the 0C isotherm does not exist everywhere, but it is still a > different variable to the one requested. You could try to persuade LS3MIP to > use the variable you are describing, but I'm not yet convinced that we can't > provide a means of describing the variable they want. > > > You are defining a variable X representing an isotherm which may lie below > frozen soil or below thawed soil. They want a variable Y which only > represents the isotherm below thawed soil. Y is a masked version of X -- but > to describe that masking in CF we would need a new area type to indicate > areas for which the surface soil temperature is above zero. > > > regards, > > Martin > > > ________________________________ > From: CF-metadata <[email protected]> on behalf of Jonathan > Gregory <[email protected]> > Sent: 18 May 2018 08:47 > To: [email protected] > Subject: Re: [CF-metadata] Standard names for LS3MIP: 8 temporal changes + 1 > feature depth > > Dear Martin > > I agree that the quantity doesn't exist everywhere, but that's the same if you > give it a name of its own rather than the more general name of depth. I > suggest > that specifying the coordinate as soil_temperature specifically implies that > it must be non-existent where 0degC is above ground or there is no ground. The > value to be given in that case needs a convention, though not necessarily as > part of the standard name definition; it could be a CMIP6 convention. It's > like > the non-existent thickness of sea ice in ice-free sea or land areas. > > Best wishes > > Jonathan > > ----- Forwarded message from Martin Juckes - UKRI STFC > <[email protected]> ----- > > > Date: Thu, 17 May 2018 17:26:58 +0000 > > From: Martin Juckes - UKRI STFC <[email protected]> > > To: Jonathan Gregory <[email protected]>, "[email protected]" > > <[email protected]> > > Subject: Re: [CF-metadata] Standard names for LS3MIP: 8 temporal changes + 1 > > feature depth > > > > Dear Jonathan, > > > > > > Yes, it should be 0C, not 0K. > > > > > > I don't think the approach you suggest will work because what we need is > > the depth of the first 0C isotherm assuming surface temperature above 0C. > > We don't want the depth of the 0C isotherm in regions where the surface > > temperature is < 0C. I can't see any way to include these conditions in > > existing CF attributes, can you? > > > > > > regards, > > > > Martin > > > > > > > > ________________________________ > > From: CF-metadata <[email protected]> on behalf of Jonathan > > Gregory <[email protected]> > > Sent: 17 May 2018 17:30 > > To: [email protected] > > Subject: [CF-metadata] Standard names for LS3MIP: 8 temporal changes + 1 > > feature depth > > > > Dear Martin > > > > All the change_over_time ones look fine to me, thanks. > > > > > 2.1 dmlt Depth to soil thaw [m] (CliC) > > > Depth from surface to the zero degree isotherm. Above this isotherm T > > > > 0o, and below this line T < 0o. > > > > > > When the surface temperature is above 0K and there is frozen soil at some > > > point beneath the surface, thawed_soil_depth is the distance from the > > > surface to the first 0K isotherm. When there is no thawed soil layer, the > > > parameter should be reported as missing. > > > > > > + Proposed: thawed_soil_depth > > > > Could we use the standard_name of depth for this, with a coordinate > > variable of soil_temperature=0degC? > > (I think 0degC is intended above, not absolute zero) > > > > Best wishes > > > > Jonathan > > _______________________________________________ > > CF-metadata mailing list > > [email protected] > > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > > ----- End forwarded message ----- > _______________________________________________ > CF-metadata mailing list > [email protected] > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata ----- End forwarded message ----- _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
