[MITgcm-support] build errors for llc_1080

[Ryan Abernathey]

Do you recommend using Crank-Nickelson?



On Wed, Nov 16, 2016 at 4:28 PM, Ryan Abernathey <ryan.abernathey at gmail.com>
wrote:

> Thanks Dimitris, this is just what I was looking for.
>
> -R
>
> On Wed, Nov 16, 2016 at 4:05 PM, Dimitris Menemenlis <
> dmenemenlis at gmail.com> wrote:
>
>> yup.  initially you need to run with small time step, until you get rid
>> of initialization shock.
>>
>> here is a spin-up history for llc1080,
>> but note that first year is without tides
>>
>> llc1080 initial conditions and spin-up:
>> - start on January 1, 2010
>>   from Hong's 2009-2011 3-year CS510 adjoint-method estimate.
>> - run for one year with ERA-interim, with corrected dlw.
>> - kick in tides, atmospheric pressure forcing, and 0.14-degree
>>   atmospheric analysis on January 1, 2011
>>
>> run_day1, deltaT=30, dumpFreq=3600, ERAi, January 1, 2010
>> run_day2_10, deltaT=90, dumpFreq=21600, ERAi, January 2-10, 2010
>> run_day10_20, deltaT=120, dumpFreq=21600, ERAi, January 11-20, 2010
>> run_day20_45, deltaT=180, dumpFreq=21600, ERAi, January 21 - February 14,
>> 2010
>> run_day45_90, deltaT=240, dumpFreq=86400, ERAi, February 15 - March 30,
>> 2010
>> run_day135,  deltaT=240, dumpFreq=86400, ERAi, March 31 - May 16, 2010
>> run_year1, deltaT=240, dumpFreq=86400, ERAi, May 17, 2010 - March 16, 2011
>>
>> run_EOG, deltaT=240, dumpFreq=86400, ECMWF 0.14-deg analysis, January
>> 1-20, 2011
>>
>> run_EOG_pres, deltaT=240, dumpFreq=86400, ECMWF + atmos pressure, January
>> 1-16, 2011
>>
>> run_EOG_press_tides, deltaT=120, dumpFreq=86400, ECMWF + pressure +
>> tides, January 1-20,  2011
>>
>> run_cn_dt120_crash, deltaT=120, dumpFreq=86400, ECMWF + pressure + tides
>> + Crank-Nickelson, January 1-5,  2011
>> run_2011, deltaT=90, dumpFreq=86400, ECMWF + pressure + tides +
>> Crank-Nickelson, January 5 - December 31,  2011
>> run_day732_896, deltaT=90, dumpFreq=3600, ECMWF + pressure + tides +
>> Crank-Nickelson, January 1 - June 20, 2012
>>
>>
>> Dimitris Menemenlis
>>
>> On Nov 16, 2016, at 11:59 AM, Ryan Abernathey <ryan.abernathey at gmail.com>
>> wrote:
>>
>> Thanks again for the advice. I got it to start timestepping!
>>
>> ...Unfortunately I had NaN's by timestep 2... ;p
>>
>> My guess is that I have to reduce my timestep (it's 90 in the example)
>> for a few iterations to overcome the geostrophic adjustment shock, as with
>> most spinup. However, before embarking on this, since people (Dimitris)
>> have already been through the procedure, I thought I would ask for your
>> guidance.
>>
>> -Ryan
>>
>>
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20161116/22065fc1/attachment.htm>