[MITgcm-support] Delaying diagnostics, gchem

[Jean-Michel Campin]

Hi Andrew,

Just a comment to add to Martin's reply, see below.

On Wed, May 08, 2019 at 04:12:01PM +0200, Martin Losch wrote:
> Hi Andrew,
> 
> the diagnostics does its first output after 
> "timePhase + multiple of |frequency|???
> (you can find a description in almost any ???data.diagnostics??? in the verification directory)
> i.e., you just specify for your nth-stream
> timePhase(n) = 31104000.,
> and then the model will start writing this stream after 31104000sec (360days). I am not sure what the first record is, but then it will do the ???frequency(n)??? iterval.
> 
> For gchem, there is nothing like this implemented (as far as I know), but you could easily do that by calling bling (in gchem_forcing_sep) only if a certain time has passed. You can use the placeholder runtime parameters gchem_int* or gchrm_rl* for that.
> 
> Probably you don???t want the passive tracers to move, either (expensive), so you???d have to do something there, too, but that???s probably more involved.

The "late starting" feature is already implemented for ptracers but not yet for gchem
(and pkgs called within gchem).
In data.ptracers, you will need to specify a "starting time":
 PTRACERS_startStepFwd(n)= {this_ptr_starting_time_in_seconds},
for each tracer id number "n"

Cheers,
Jean-Michel

> 
> Martin
> 
> > On 8. May 2019, at 15:56, Andrew Twelves <agt_94 at hotmail.co.uk> wrote:
> > 
> > Hi all,
> > 
> > My question: is there a simple way at the start of a run to prevent diagnostics from writing until a set iteration has been past, and similarly to suppress gchem (specifically BLING) from operating until the physical model is spun up? (other than just stopping the submission and restarting with pickup files)
> > 
> > Many thanks,
> > 
> > Andrew Twelves
> > University of Edinburgh
> > _______________________________________________
> > MITgcm-support mailing list
> > MITgcm-support at mitgcm.org
> > http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support