[MITgcm-devel] changes in pkg/layers

[Jean-Michel Campin]

Hi Ryan,

Would this work to just make the 1rst mofification (prho shifted by
-1000), which involves very few lines of code, and leave the 2 other
modifs for later, after you sort out your changes ?

The reason beeing that, since there is this new diagnostics, changing
prho now seems a good time.

Cheers,
Jean-Michel

On Tue, May 12, 2015 at 11:29:20AM -0400, Ryan Abernathey wrote:
> Jean-Michel,
> 
> These all sound like good suggestions. The problem is that I have made some
> major modifications and feature enhancements to layers which I have not yet
> managed to check in. (A few months ago I emailed you about the CVS problems
> I encountered, but I haven't managed to get around this yet.)
> 
> Can I ask you to wait a few days to do these changes to allow me to first
> (finally) merge the changes I have been working on? What would be the best
> way to deal with this situation?
> 
> Thanks,
> Ryan
> 
> On Tue, May 12, 2015 at 11:10 AM, Jean-Michel Campin <jmc at ocean.mit.edu>
> wrote:
> 
> > Hi,
> >
> > I would like to make little changes in pkg/layers:
> > 1) substract 1000 to "prho" = the potential density
> >   that pkg/layers uses as tracer field.
> >  motivation:
> >   a) when I output "prho" using 32-bit precision file,
> >     the -1000 shift would save at least 2 digits of precision.
> >   b) this is a common practise in oceanography to use rho-1000
> >     variable for potential density.
> >   c) since I've just added (yesterday) a diagnostics for this
> >     prho field (+ bring back the snap-shot output of prho that
> >     has been missing for long time when density-layers is not in
> >     first position in the "layers_num" list), this could be a good
> >     time to make this modification.
> > 2) add few check and stop regarding parameter settings, especially:
> >   a) checking for inconsistency between "layers_name" and layers_num
> >    value.
> >   b) if mixing old setting (LAYER_nb, layers_kref, layers_G, useBOLUS)
> >   and new settings (layers_num, layers_krho, layers_bounds, layers_bolus).
> > 3) move layer diagnostics array out of commom blocks and define them
> >   as local variables in layers_calc.F
> >  (outside common blocks, the missing re-init of layer non-weighted velocity
> >   and layer probability - fixed yesterday - would have be caught by some
> >   "-devel" compiler option).
> >
> > Comments ? suggestions ?
> >
> > Cheers,
> > Jean-Michel
> >
> > _______________________________________________
> > MITgcm-devel mailing list
> > MITgcm-devel at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-devel
> >

> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel