[MITgcm-support] How to pickup when using pkg/cal?

Martin Losch martin.losch at awi.de
Tue Mar 28 08:52:24 EDT 2017


Hi DImitris,

can we repoduce this in a simple set up (e.g. by modifying one of the verification experiments)?

Martin

> On 28 Mar 2017, at 02:59, Dimitris Menemenlis <dmenemenlis at gmail.com> wrote:
> 
> I just ran across the same wonderfully informative error message:
>   "cal_Set: The specifications are incomplete. Please refer
>    to the documentation.”
> that Alistair encountered in 2004 (see attached exchange
> between Alistair and Patrick).  One additional clue is that while
> "niter0=262980.,” causes this bug to show up, the model starts
> fine with "niter0=262944.,”.  Anybody else has encountered this
> bug and/or has an idea how to fix more permanently in pkg/cal?
> 
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
> Hi P.
> 
> data.cal:
>   &CAL_NML
>   TheCalendar = 'gregorian',
>   startDate_1 = 19990801,
>   startDate_2 = 000000,
>   &
> 
> data.exf:
>   &EXF_NML
> #
>   hfluxstartdate1   = 19990731,
>   hfluxstartdate2   = 000000,
>   hfluxperiod       = 21600.,
> #
>   atempstartdate1   = 19990731,
>   atempstartdate2   = 000000,
>   atempperiod       = 21600.,
> ...
> 
> data:
>   &PARM03
>   nIter0=3600,
>   endTime=432000,
>   deltaTmom=60.,
> 
> Thx,
> 
> A.
> 
> Patrick Heimbach wrote:
> >
>  Hi Alistair,
> 
> >
>  
> 
> >
>  I use pickups with pkg/cal rather frequently,
> 
> >
>  so it seems to work sometimes.
> 
> >
>  How about you give more info, e.g. contents
> 
> >
>  of your data, data.cal, data.exf files.
> 
> >
>  
> 
> >
>  Suspicion is your data.cal is somehow screwed up.
> 
> >
>  cal_set tries to combine
> 
> >
>  startdate_1 (YYYYMMDD), startdate_2 (HHMMSS) specifications 
> 
> >
>  with starttime, endtime (or nTimeSteps) to determine run time interval
> 
> >
>  (in seconds).
> 
> >
>  
> 
> >
>  -Patrick
> 
> >
>  
> 
> >
>  
> 
> >
>  
> 
> >
>    > -----Original Message-----
> 
> >   > From: mitgcm-support-bounces at mitgcm.org
>  
> 
> >   > [mailto:mitgcm-support-bounces at mitgcm.org
> ] On Behalf Of 
> 
> >
>    > Alistair Adcroft
> 
> >
>    > Sent: Wednesday, December 01, 2004 9:22 AM
> 
> >
>    > To: MITgcm support
> 
> >
>    > Subject: [MITgcm-support] How to pickup when using pkg/cal?
> 
> >
>    > 
> 
> >
>    > 
> 
> >
>    > I'm trying to pickup (restart) a run that's using pkg/cal 
> 
> >
>    > and I get the 
> 
> >
>    > error message
> 
> >
>    > 
> 
> >
>    >   "cal_Set: The specifications are incomplete. Please refer
> 
> >
>    >    to the documentation."
> 
> >
>    > 
> 
> >
>    > It's the sort of helpful error message that I would write 
> 
> >
>    > but it's not 
> 
> >
>    > me this time. I checked the pelican documentation 
> 
> >
>    > (ECCO/cal) and there 
> 
> >
>    > is no mention of the error nor how to do a pickup. Can 
> 
> >
>    > someone post the 
> 
> >
>    > answer?
> 
> >
>    > 
> 
> >
>    > A.
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support




More information about the MITgcm-support mailing list