[MITgcm-support] MITgcm-support Digest, Vol 98, Issue 16

[Daiwei (David) Wang]

Hi Zhaomin,

I'd first check where the runs blew up. Is it near cubic corners where 
the horizontal spacings are small? Narrow bays, fjords? Since it seems 
independent of the length of time step, chances are you are heating or 
cooling some grid cells too much before the heat can be dispersed by 
advection and diffusion.

David

On 8/25/11 12:41 PM, Wang, Zhaomin wrote:
> Hi Jean-Michel,
>
> Similar to ecco2 project, we use cube sphere 510 configuration. (All forcing files and initial conditions and namelist files are downloaded from JPL.) The only difference is that we use yearly constant forcing. For example, we use repeated 1992 forcing, so there is no interannual variability. In the run with previously reported problem, we used 1992 winds and 1998 buoyancy forcing for sensitivity studies. The model could be run for about 10 years before the instability problem. We also found similar problem for runs with other kind of forcing. For example, in the run with 1992 constant forcing, the model can only be run for about 31 year. Usually, if we reduce deltaT from 1200s to 600s, then it is Ok. But for the run using 1992 winds and 1998 buoyancy forcing, I have reduced deltaT to 300s and still have the problem. So I wondered if there were another things that caused the problem.
>
> Please let me know if you need further details.
>
> Thanks.
> Zhaomin
>
> -----Original Message-----
> From: mitgcm-support-request at mitgcm.org [mailto:mitgcm-support-request at mitgcm.org]
> Sent: 25 August 2011 17:00
> To: mitgcm-support at mitgcm.org
> Subject: MITgcm-support Digest, Vol 98, Issue 16
>
> Send MITgcm-support mailing list submissions to
> 	mitgcm-support at mitgcm.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://mitgcm.org/mailman/listinfo/mitgcm-support
> or, via email, send a message with subject or body 'help' to
> 	mitgcm-support-request at mitgcm.org
>
> You can reach the person managing the list at
> 	mitgcm-support-owner at mitgcm.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of MITgcm-support digest..."
>
>
> Today's Topics:
>
>     1. Re: time step problem for cs510? (Jean-Michel Campin)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 24 Aug 2011 22:36:39 -0400
> From: Jean-Michel Campin<jmc at ocean.mit.edu>
> To: mitgcm-support at mitgcm.org
> Subject: Re: [MITgcm-support] time step problem for cs510?
> Message-ID:<20110825023639.GB12672 at ocean.mit.edu>
> Content-Type: text/plain; charset=us-ascii
>
> Hi Zhaomin,
>
> May be a little bit of 'context' type information might help
> (type of set-up, type of problem you want to addess, ...).
> And also some numerical details about your set-up could
> also be useful (in general, the usefulness of the response you
> could expect from this list is largely function of the
> how precise the problem is described).
>
> Cheers,
> Jean-Michel
>
> On Wed, Aug 24, 2011 at 03:19:04PM +0100, Wang, Zhaomin wrote:
>> Hi,
>>
>> We have one run probably stopped by numerical instability. We used cube sphere 510 configuration, and initially used deltaT=1200. When the run was stopped, we reduced deltaT to 600. When the one was stopped again, we used deltaT=300. However, the run was also stopped even for this very small time step. We did not try to reduce deltaT further, because the instability may not be caused by a too large time step.  We got the following error message:
>>
>> ***********************
>> WARNING: r*FacW<  hFacInf at       8 pts : bi,bj,Thid,Iter=   1   1   1   1694086
>>   e.g. at i,j=  24  24 ; rStarFac,H,eta =       NaN  4.281800E+02           NaN -1.058899E+00
>> STOP in CALC_R_STAR : too SMALL rStarFacW !
>> WARNING: r*FacS<  hFacInf at       8 pts : bi,bj,Thid,Iter=   1   1   1   1694086
>>   e.g. at i,j=  23  25 ; rStarFac,H,eta =       NaN  4.281800E+02           NaN -8.253981E-01
>> STOP in CALC_R_STAR : too SMALL rStarFacS !
>> WARNING: r*FacC<  hFacInf at       5 pts : bi,bj,Thid,Iter=   1   1   1   1694086
>>   e.g. at i,j=  23  24 ; rStarFac,H,eta =       NaN  4.281800E+02           NaN
>> STOP in CALC_R_STAR : too SMALL rStarFacC !
>> ********************
>>
>> Thanks.
>>
>> Zhaomin
>>
>> British Antarctic Survey
>> High Cross, Madingley Road
>> Cambridge, CB3 0ET
>> UK
>> Tel: +44(0)1223221488
>> Fax: +44(0)1223221279
>>
>> -- 
>> This message (and any attachments) is for the recipient only. NERC
>> is subject to the Freedom of Information Act 2000 and the contents
>> of this email and any reply you make may be disclosed by NERC unless
>> it is exempt from release under the Act. Any material supplied to
>> NERC may be stored in an electronic records management system.
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
> ------------------------------
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
> End of MITgcm-support Digest, Vol 98, Issue 16
> **********************************************
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support